THE PLANT WORIvD 241 



invertebrates and in land molluscs, as well as in the general problems of 

 the growth of coral reefs. His observations led him to the discovery that 

 pelagic animals are much more abundant in regions of great ocean cur- 

 rents and near the coasts of large islands and continents than far from 

 land, and that animals which swim at the surface at night sink deep 

 below the surface in the daytime. 



Dr. Mayer told me that distinct terraces could be seen from the sea- 

 ward in the cliffs bordering the east coast of the northern portion of the 

 island and that there were caverns at a considerable elevation above the 

 level of the sea which must have been worn by the waves before the 

 upheaval of the island to its present height. I had heard of several of 

 these " liyang,'' which the natives say are frequented by bats, but I have 

 not yet visited them. One thing which makes me hesitate to place implicit 

 confidence in the theory of submarine erosion is the fact that there are 

 comparatively few reefs on the east or windward side of the island, which 

 is most subjected to the action of the waves and currents ; while on the 

 west, or lee side, occur broad coral flats as level as a floor, bounded by 

 a barrier reef and covered by very shallow water. The elevation of 

 the island a few feet would convert these flats into a new terrace, and in 

 a few years it would become covered with vegetation and resemble the 

 terraces already existing on the island. Mr. Agassiz believes that many 

 of the interior lagoons or sounds have been eroded by submarine action 

 from solid coral. Here in Guam we have an ancient lagoon which has 

 become a marsh (/a Cienaga) almost choked up by reeds and other veg- 

 etation and enclosing a number of islets on which coconuts are growing. 

 Instead of being deepened by erosion, this lagoon is gradually becoming 

 filled up. In excavating ditches for draining in the cienaga my men 

 recently came upon coral a few feet below the general surface of the 

 marsh. Moreover, the depths of such harbors as that of Apia, in Samoa, 

 and of San I^uis de Apra, in Guam, are so great and the coral growth 

 seems to be so constantly encroaching upon them that I cannot believe 

 they were excavated by erosion. It seems quite possible, however, that 

 there are coral atolls which have been formed by the dissolving of 

 limestone or dead coral in the center while the live coral of the periphery 

 continues to grow. It is interesting to note the occurrence of marine 

 molluscs and of fishes in lagoons entirely enclosed by a rim of coral. I 

 have already recorded the bubbling forth of fresh-water springs i^bobo) near 

 the line of high water and even from below the surface of the sea near the 

 coast of this island, in the vicinity of Tutuhon. These are probably the 

 outlets of subterranean drains from sinks ilupog) in the interior of the 

 island. 



When we reached home we found that Susana's nephews had been 

 collecting specimens of our fresh- water fishes for Mr. Agassiz. I think 



