Feb. 28, 1889] 



NATURE 



427 



reefs and islands — such are Foraminifera, Sponges, Polyzoa, 

 Annelids, Echinoderms, and Calcareous Algzc. The relative 

 proportions of these difTerent organisms in a reef vary with the 

 region, with the depth, and with the tenipernture, but members 

 of what are known under the term of reef corals appear always 

 to predominate. 



The animals of the true reef-building species resemble the 

 common sea-anemones in structure and size ; the individual 

 Polyps may vary from the eighth of an inch in diameter to over 

 a foot. Some of the structures built by colonies may exceed 20 

 feet in diameter. 



There may be great variety in the appearance of submerged 

 reefs as they rise from banks of a different nature, form, and 

 extent, as, indeed, was pointed out long ago by Chamisso. There 

 may be differences due also to the kinds and abundance of deep- 

 sea animals living on such banks, as well as differences due to 

 currents, temperature, and other meteorological conditions. 



From the very first the plantations situated on the outer edge 

 will have the advantage, from the more abundant supply of food 

 and the absence of sand in the water, which last more or less 

 injuriously affects those placed towards the interior. Chamisso 

 attributed the existence of the lagoon to the more vigorous 

 growth of the peripherally situated corals of a reef, as compared 

 with those placed towards the middle, and in this he was to a 

 large extent right, but the symmetrical form of the completed 

 atoll is chiefly due to the solution of the dead carbonate of lime 

 structures. The Great Chagos Bank illustrates the irregular way 

 in which such a large bank of coral plantations approaches the 

 surface. When these, however, reach the surface, they assume 

 slowly a more regular outline, those on the outer edge coalesce, 

 and ultimately form a complete ring of coral reef, and the lagoon 

 becomes gradually cleared of its coral patches or islands, for, as 

 the atoll becomes more perfect, the conditions of life within the 

 lagoon become less and less favourable, and a larger quantity of 

 dead coral is removed in solution. 



The coral atoll varies greatly in size and form : it is usually 

 more or less circular, horse-shoe shaped, and may be one or 

 over fifty miles in diameter. The breakers spend their fury on 

 the outer edge, and produce what is known as the broad shore 

 platform ; but within, trees descend to the very shore of the 

 lagoon, where there is quiet water, and a ship may often enter 

 on the lee side of the atoll and find safe anchorage. 



In this connection it is important to bear in mind the relation 

 which exists between the periphery and the superficial area of 

 the lagoon in atolls of different sizes. If the coral plantations 

 which rise from the top of a submerged mountain have an area 

 of one square mile, then on reaching the surface of the waves 

 there will be a shallow depression in the centre owing to the more 

 rapid growth of the outer edge. Such an atoll will have, if it 

 be a square, four miles of outer reef for the supply of coral sand 

 and other debris, and these being washed and blown into the 

 one square mile of shallow lagoon it is likely to become filled 

 up, the result being a small island with dry lagoon, in which may 

 be found deposits of sulphate of lime, magnesian and phosphatic 

 rocks, and guano — all these testifying to the great age of the 

 island and absence of subsidence in the region. It is only atolls 

 with a diameter of less than two miles that thus become filled 

 up. In other and larger plantations, rising from a more ex- 

 tensive bank, the conditions are very different. In this larger 

 atoll — say four miles square — there is now only one mile of outer 

 reef to each square mile of lagoon, instead of four miles of 

 outer reef to the one square mile of lagoon in the smaller atoll. 

 Only one-fourth of the detrital matter and food enters the larger 

 lagoon, from the outside, per square mile of lagoon, and hence 

 there is proportionately less living coral, the solvent agencies 

 predominate, and the lagoon is widened and deepened. Growing 

 seawards on the outer face and dissolving away in the lagoon, 

 the whole expands after the manner of a fairy ring, and the 

 . ibbon of reef or land can never in consequence increase beyond 

 a half or three-quarters of a mile in width, it being usually much 

 less. I have recently made a very careful comparison of the 

 latest Admiralty Survey of the lagoon of Diego Garcia with the 

 one made many years ago, and the result appears to me to 

 indicate that the area of the lagoon has considerably increased in 

 the interval, and the average depth is a little greater than 

 formerly, although shallower in some places. 



Atolls may occur far away from any other land, but it more 

 frequently happens that they are arranged in linear groups, in 

 this respect resembling volcanic islands. Extensive banks may 

 be crowded with small atolls, like the Northern Maldives ; or a 



bank may be occupied by one great and perfect atoll twenty to 

 forty miles in diameter, like some of the Southern Maldives and 

 the Paumotus. In some instances the large atolls appear to have 

 resulted from the growth and coalescence of the smaller marginal 

 atclls ; especially does this seem to have been the case with the 

 large Southern Maldives. 



'ihe outer slopes vary greatly in different reefs, and in different 

 parts of the same reef. When there is deep water beyond, the 

 reef very often extends out with a gentle slope to a depth of 25 

 to 40 fathoms, and is studded with living coral, the bosses and 

 knobs becoming larger in the deeper water farthest from the 

 reef, where there are great overhanging cliffs, which eventually 

 fall away by their own weight, and form a talus on which the reef 

 may proceed further outwards. Occasionally there is a very 

 steep descent almost at once from the outer edge. Thus, the 

 deeper the water beyond, the more slowly will the reef extend 

 seawards. In reefs with a very gentle slo|ie outside, the corals 

 are frequently overhanging at depths of 6 or 7 fathoms, for in 

 these instances the lower part of the sea-face of the reef 

 is rendered unsuitable for vigorous growth, in consequence 

 of the sand which is carried in by waves coming over the 

 comparatively shallow depths outside ; in these cases, lines of 

 growing corals, or a submerged barrier, are sometimes met 

 with in deep water some distance seawards from the edge of 

 the reef. 



As has been stated, the lagoon in many of the smallest atolls has 

 been filled up, but this never appears to happen in atolls with a 

 diameter of over two miles unless there be distinct evidence of • 

 upheaval. In perfectly-formed atolls — that is, those in which the 

 reefs are nearly continuous throughout — the deepest water is 

 found towards the centre of the lagoon, and there is a relation 

 between this depth and the depth of water beyond the outside 

 reefs. In North and South Minerva reefs, in the South Pacific, 

 where the outside depths are very great, there are depths down 

 to 17 fathoms in the lagoons, which are apparently clear of 

 coral heads. Here we may suppose that the central parts of the 

 lagoon have for a long time been exposed to the solvent action 

 of sea-water, owing to the slow lateral growth of the reef as a 

 whole. In the same regions the Elizabeth and Middleton reefs, 

 which are about the same size, have only 4 or 5 fathoms within 

 the lagoons, and the depths outside the reefs are, at the distance 

 of a mile, mostly within the lOO-fathom line, and sometimes 

 less than 50 fathoms. There are also many coral heads within 

 the lagoons. Here we may suppose the atolls to be more recent, 

 and to have extended more rapidly than in the case of the 

 Minerva reefs. If the depths beyond the reefs be taken into 

 consideration, then there is usually a direct relation between the 

 depth of the lagoon and its diameter. The greatest depths, even 

 in the largest atolls, do not exceed 50, or at most 60, fathoms ; 

 they are usually much less. In atolls which are deeply sub- 

 merged, or have not yet reached the surface, which have wide and 

 deep openings into lagoon-like spaces, this relation may not exist. 

 In these instances the secretion and deposition of carbonate of 

 lime may be in excess of solution in all parts of the lagoon. It 

 is only when the atoll reaches the surface, becomes more perfect, 

 and its lagoon waters consequently less favourable to growth, 

 that the solution of the dead corals and calcareous rtt'i!^r/^ exceeds 

 any secretion and deposition that may take place throughout the 

 whole extent of the lagoon ; it is then widened and deepened, 

 and formed into a more or less perfect cup-like depression, unless 

 the lagoon be of small size and is filled up. 



The whole of a coral reef is permeated with sea-water like a 

 sponge ; as this water is but slowly changed in the interior parts, 

 it becomes saturated, and a deposition of crystalline carbonate 

 of lime frequently takes place in the interstices of the corals 

 and coral dSris. In consequence of the solution of coral debris 

 and the re-deposited lime occupying less space, large cavities are 

 formed, and this process often results in local depressions in 

 some islands, as, for instance, in Bermuda. At many points 

 on a reef where evaporation takes place there is a deposition 

 of amorphous carbonate of lime cementing the whole reef 

 materials into a compact conglomerate-like rock. .— r^* 



The fragments of the various organisms broken off from the 

 outer edge during gales or storms are piled up on the upper 

 surface of the reef, and eventually ground into sand, the result 

 being the formation of a sandy cay or shoal at some distance 

 back from the outer edge of the reef— the first stage in the 

 formation of dry land. 



The fragments of pumice thrown up into the ocean during 

 far-distant submarine eruptions, or washed down from volcanic 



