THE BIOLO&Y OF THE FUNAFUTI ATOLL AND REEF FORMATION. 137 



fact that a dead specimen of Distichopora rosea was collected by him on the beach 

 of Funafuti. 



(4) Foraminifera. The remains of these Rhizopoda are present in all the beach 

 sands, as well as in the lagoon sediment in close proximity to the shores. Pieces 

 detached from any part of the reef platforms show them to have taken an important 

 part in the formation of the reef rock. The somewhat difficult task of ascertaining 

 the distribution of this factor in reef formation was not attempted by me. Mr. 

 Frederick Chapman's papers supply detailed information upon this point.* 



(4) Mode of Occurrence of the Chief Reef -Forming Organisms. 



The order of the previous section is here followed. (1) Lithothamnion. Of the 

 three forms occurring, two are distinctly lichenous ; the third is fruticose, growing 

 erect in isolated pieces, and branching in a shrub-like manner. The last-named has 

 therefore not the binding effect of the other two forms, and, as a reef- forming organism, 

 must be classed with the corals, playing, like them, a passive part. The lichenous 

 Lithothamnion and the knobby form, on account of their incrusting habits, play a 

 very important part in the formation of the reef Being purely incrusting, they 

 require for their growth a foundation over which they can spread. The limit of such 

 foundation is the limit of their extension. For further growth the presence of addi- 

 tional foundation is required. Should such be present in very close proximity to the 

 original one, the Lithothamnion would reach across to it and continue its incrusting 

 work, neighbouring masses being by this means firmly united together. The 

 Lithothamnion by no means confines itself to lifeless material, everything that crosses 

 its path is overgrown and overpowered. Excellent illustrations of this are afforded 

 by the Pocillopora corals on the ocean platform of the Island of Funafuti already 

 mentioned. The separate branches of this coral are widely separated ; the nullipore 

 does not attack it at the basis of the branches and grow upwards, but expands over 

 the surface of the coral from tip to tip of the branches, leaving hollow spaces under- 

 neath. These spaces eventually fill up, so that part of the coral which has been killed 

 by the overgrowing Lithothamnion becomes a solid mass. Whether this takes place 

 by infiltration or by precipitation, or by the actual increase in thickness of the 

 Lithothamnion, is a geological question. The thickness of an individual crust of 

 Lithothamnion noticed at Funafuti varies from a covering too thin to be detected at 

 its edges even by touch, to a layer of 1 inch in thickness. A satisfactory determination 

 was not arrived at as to whether growth was taking place in thin layers on the surface 

 only, or whether the whole thickness was actually living. Certain observations, 

 however, to be quoted in the sub-section (6) on the " Rate of Growth," point to the 

 former method of growth. 



(2) Halimeda. In regard to the mode of occurrence of these algse, the roundish - 



* 'Linnean Soc. Journal,' Zoology, vol. 28 (1900-03), pp. 1, 161. 



T 



