424 



NATURE 



[Fed. 



vertical circle with it, while for the minimum both stars must be 

 at the same altitude. M. Loewy then finds that the greatest 

 variation of refraction will be obtained with an angle of 30° for 

 the double mirror, but as with this there would be (for the 

 latitude of Paris) a minimum interval of 6h. 35m. between the 

 two epochs of observation, he prefers to take an angle of 45° for 

 the double mirror, sacrificing only 15" in the effect of refraction, 

 while reducing the interval between the observations to 4h. 44m. 

 This is the minimum value of the interval found by selecting 

 the pair of stars so that their common zenith distance at the 

 second epoch is equal to the angle of the double mirror, or half 

 the angular distance between the two stars. 



The geometrical conditions thus found by M. Lcewy to give 

 the maximum effect in the minimum interval of time between 

 the observations may be somewhat modified in practice, provided 

 the angular distance between the stars does not differ by more 

 than a few minutes from twice the angle of the double mirror. 

 M. Loewy has thus been able to find some twenty pairs of bright 

 stars suitable for the determination of refraction by this method. 

 In its practical form the method consists in the measurement of 

 the angular distance between a pair of stars 90° apart when one 

 of the stars is near the horizon and the other near the zenith, 

 and again when both the stars are at about the same altitude. 

 It is not necessary that at the former epoch the low star should 

 be very near the horizon, for, as M. LcEwy points out, observa- 

 tions may be advantageously continued till the altitude is nearly 

 20°, and thus the constant of refraction may be determined from 

 observations which are practically unaffected by any uncertainty 

 in the law of refraction. 



It will readily be understood that the observation of the low 

 star may be made either when it is rising or when it is setting. 

 In the latter case the observation of the stars at equal altitude 

 would precede that for which one of the stars is setting. By 

 combining the observations of two pairs of stars chosen so that 

 the first pair is rising when the effect of refraction on the second 

 is a minimum, and that the first pair is at the minimum when 

 the second pair is setting, the influence of any change in the 

 angle of the double mirror will be eliminated by taking the 

 mean of the two determinations, while the difference of these 

 will give four times the change of angle in the interval, thus 

 affording a precise determination of any such change, if it 

 exists. 



Various other methods are proposed by M. Loewy for deter- 

 mining the refraction at any altitude without assuming its law of 

 variation. These methods, however, appear to involve practical 

 difficulties, as they either assume the absence of irregular varia- 

 tions in the refraction at an altitude of 10°, or require the 

 construction of several double mirrors with different angles. 

 They may be considered as supplementing the first method ; and 

 they are of interest as giving a direct measure of refraction 

 independently of any theory. 



The practical determination of the constants of aberration and 

 refraction by the new method is being carried out by M. Lcewy 

 and M. P. Puiseux with the equatorial co7ide of the Paris Obser- 

 vatory, and the series of observations made during the past 

 twelve months confirms in the most satisfactory manner the 

 theoretical conclusions. M. Loewy finds that the variations of 

 the distances are really free from systematic errors, and he con- 

 siders that the constant of refraction will be more accurately 

 determined from a few nights' observations with his new method 

 than from years of meridian observations. 



In conclusion, I can only allude in the briefest terms to the 

 other important researches for which astronomers are indebted 

 to M. Loewy. The following is a summary of the other new 

 methods of instrumental research which M. Loewy has devised 

 in the last few years : — 



(i) A method for determining the flexure of transit-circles at 

 various zenith distances by means of an optical apparatus inserted 

 in the central cube. This has been used to find the flexure of 

 two transit-circles at the Paris Observatory, the absolute values 

 of the flexure for the two ends of the telescope and for the axis 

 being independently determined. 



(2) A method for obtaining the latitude without making use of 

 the declinations of fundamental stars. 



(3) A general method for determining right ascensions without 

 relying on assumed right ascensions of polar stars. 



(4) A method for finding on each night the absolute declina- 

 tions of stars without the necessity for observations of polar 

 stars at upper and lower transit. 



(5) Methods for determining directly the two co-ordinates of 



polar stars without a previous investigation of the instrumental 

 errors. 



All these methods except the first are based on the observation 

 of close circumpolar stars in R.A. and N.P.D. out of the 

 meridian at various points of the circles described by them. 

 Conjugate observations either of a single star or of a pair ot 

 stars having the same N.P.D. are made with a transit-circle, 

 havmg a field of view of 2°, at equal intervals (about two hours) 

 before and after meridian passage or before and after passacre 

 over the hour-circle of 6h. east or west. The special methods 

 of observation are developed in a series of communications to the 

 French Academie des Sciences made in the years 1883 and X885, 

 and during the last two years M. Renan has applied these new 

 methods to a determination of the latitude of the Paris Observatory 

 based on eighty very accurate results. 



The account which I have given of M. Loewy's inventions and 

 researches is necessarily very imperfect, and I have had to pass 

 over many points of interest in the application of his methods. 

 But I trust that the summary I have made will at any rate 

 suffice to show the very high importance of M. Loewy's labours, 

 and that they fully deserve the recognition which is to-day 

 given to them, whether we have regard to the originality of the 

 methods or to the value of the results which are to be obtained 

 from them. 



STRUCTURE, ORIGIN, AND DISTRIBUTION 

 OF CORAL REEFS AND ISLANDS.^ 



"pHE picturesque beauty of the coral atoll, seated 'mid a waste 

 of troubled waters, with its circlet of living green, its quiet, 

 placid lagoon, and its marvellous submarine zoological gardens, 

 has long been celebrated in the descriptions of voyagers to 

 tropical seas. The attempt to arrive at a correct explanation of 

 the general and characteristic form and features of these reefs 

 and islands has, for an equally long period of time, exercised the 

 ingenuity of thoughtfulmen. 



Coral reefs are the most gigantic and remarkable organic 

 accumulations on the face of the earth. They are met with in 

 certain tropical regions, and are huge masses of carbonate of 

 lime, secreted from ocean waters by myriads of marine organisms. 

 While the great bulk of the reef consists of dead corals, skeletons, 

 and shells, the outer surface is clothed with a living mantle of 

 plants and animals. This is especially the case on the outer and 

 seaward face of the reef, where there are, at all times, myriads 

 upon myriads of outstretched and hungry mouths, and not the 

 least interesting questions connected with a coral reef are those 

 relating to how these hungry mouths are satisfied. 



It is to the power of these organisms of secreting carbonate of 

 lime from sea-water — building up and out generation after genera- 

 tion on their dead selves— that the coral reef owes its origin. 

 So wonderful and unique is the result, that combination for a 

 definite end has sometimes been attributed to these reef-builders. 



There is, however, another process ever at work in the ocean, 

 in a sense antagonistic to that of secretion of carbonate of lime 

 by organisms, which has much to do in fashioning the more 

 characteristic features of coral reefs. This is the solution of all 

 dead carbonate of lime shells, skeletons, and calcareous debris, 

 wherever these are exposed to the action of sea- water. As soon as 

 life loses its hold on the coral structures, and wherever these dead 

 carbonate of lime remains are unprotected by rapid accumulation 

 or crystalline depositions, they are silently, surely, and steadily 

 removed in solution. This appears to be one of the best established 

 oceanographical facts, and any theories concerning the general 

 economy of the ocean which fail to take account of this universal 

 agency are most likely to be at fault. We know something about 

 the rate of solution, probably more than we do about the rate of 

 growth and secretion of carbonate of lime by the coral Polyps. 

 It has been shown that the rate of solution varies with tempera- 

 ture, with pressure, and with the amount of carbonic acid 

 present in the water. It is on the play of these two opposing 

 forces— the one vital and the other chemical— and their varying 

 activity in different regions and under different circumstances, 

 that we rely for the explanation of many oceanographical pheno- 

 mena, especially many of those connected with oceanic deposits 

 and coral reefs. In some regions there may be more growth, 

 secretion, and deposition of shell and coral materials than solution 



^Lecture delivered by Dr. John Murray at the Royal Institution ' ore 

 Friday, March 16, 1888. Recently levised by the Author. 



