Sept., i893. THE EVOLUTION OF LIFE. 223 



live upon sea animals. The enormous numbers of terns, gulls, petrels, 

 cormorants, and so forth, are all carnivorous. Among lower forms, 

 practically only a few molluscs, echinoderms, and annelids are plant- 

 eaters, and they play an inconspicuous part in the economy of the 

 ocean. 



If a small amount of dcbyis coming from the land is left out of 

 count, the source of the food-supply of the ocean is to be found in the 

 minute organisms of the surface. Trichodesmium, Pyrocistis, Proto- 

 coccus, and the coccospheres, rhabdospheres, and diatoms form the 

 lowest link in the chain. These and the globigerinae and radio- 

 larians that feed on them, are so abundant and prolific as to supply 

 practically all the food for the animals of the ocean. Such a simple 

 pelagic food-supply Dr. Brooks believes to be, not only the funda- 

 mental, but the primaeval supply — the supply which has determined 

 the course of evolution of marine life, and secondarily of all life. 

 The conditions of pelagic life are so simple and so universal, that the 

 dawn of life appeared there. There is no fierce competition, and very 

 little stimulus for the production of diversity of habit. All the 

 metazoa have pelagic larvae, or else embryonic stages recalling such 

 larvae, and best interpreted as the degenerated vestiges of a pelagic 

 habit. During the long period in the history of the earth when there 

 was practically only pelagic life, the pelagic ancestors of the great 

 groups of the metazoa were evolved. 



These advanced little till multiplication in the number of 

 individuals drove some to the bottom or to the shore. In the strain 

 and stress of the new conditions with their more varied mutable 

 environment, increase of size and complexity, appearance of shells 

 and skeletons began. All this happened before the earliest fossiliferous 

 strata, for these show abundance of complicated types, of which their 

 minute primaeval pelagic ancestors left no traces in the sedimentary 

 rocks. Secondarily, a number of forms returned from the bottom to 

 the surface just as the Cetacea and many sea-birds returned from the 

 land to the sea, and the improved wanderers rapidly produced 

 changes in their ancestral pelagic home ; but none the less a number 

 of surface forms Dr. Brooks beheves to retain the simplest possible 

 structure, and to resemble closely the primaeval ancestors of their 

 groups. 



Among these are notably the ancestors of the Chordata. Dr. 

 Brooks criticises very closely Dohrn's famous degeneration theory of 

 the Ascidians and finds in Appendicularia a little modified ancestral 

 pelagic chordate. 



One difficulty in the way is the possible importance of metamerism ; 

 but Dr. Brooks makes light of this, and many morphologists will be 

 ready to agree with him. Metameric segmentation is a feature which 

 occurs in widely-separated groups : which appears again and again in 

 cases where there can be no possibility of community of descent. It 

 may, in fact, be said that metamerism is so characteristic of life that 



