THE RECAPITULATION HYPOTHESIS 267 



of the lateral sheets of muscle derived from the mesoblast. No 

 skull and no vertebral column are formed, the notochord 

 remaining throughout life as almost the sole representative of the 

 internal skeleton. The brain does not become distinctly 

 differentiated from the spinal cord and no paired organs of sense 

 are developed, nor does that division of the body into head and 

 trunk, which is so characteristic of typical vertebrates, take 

 place. Amphioxus is a chordate animal but it is not a true 

 vertebrate. It probably, however, represents fairly closely a 

 stage of evolution through which the ancestors of the vertebrates 

 have passed, though it becomes somewhat modified by secondary 

 features in the later stages of its development. 



It is easy enough to interpret the earlier development of Amphi- 

 oxus in terms of the recapitulation hypothesis and to recognize 

 in the ontogeny certain stages of the phylogenetic history. The 

 ontogenetic record, however, is by no means always so readily 

 decipherable. It is generally more or less obscured by secondary 

 features superposed upon it, much as an ancient picture may 

 be concealed by another painted over it. These secondary 

 features are characters which are developed in relation to the 

 requirements of the embryo itself at its various stages, for it is 

 not only the adult organism which becomes modified in structure 

 in the course of evolution but all the stages of its life-history 

 are likewise subject to adaptation. 



In the first place nourishment has to be provided for the 

 developing embryo, and this necessity is present from the very 

 first division of the fertilized ovum. The blastomeres into 

 which it divides cannot go about and seek their own food supplies 

 like so many Amoabae, they have sacrificed the power of leading 

 independent lives for the sake of remaining together and building 

 up a more or less complex multicellular body ; but none the less 

 they require feeding. Hence we find that all eggs contain a 

 larger or smaller quantity of nutrient material stored up in 

 the cytoplasm. This usually takes the form of definite granules 

 or particles of food-yolk (deutoplasm), the amount of which 

 varies immensely in different cases. If there is very little the 

 egg is said to be microlecithal as in Amphioxus; if there is 

 much it is said to be megalecithal, as in the frog and still more 

 so in the bird, and the size of the egg, as we have already had 

 occasion to point out in Chapter X. depends almost entirely upon 

 the amount of food-yolk which it contains. 



