i ORIGIN OF THE MESODERM 59 



blocks (their outer and inner walls being in contact). These sacs or 

 blocks are the mesoderm segments and their cavities are the seg- 

 im -ii tally arranged rudiments of the coelome. The subsequent fate 

 of the mesoderm segments will be traced later ((hap. IV.). 



Of the lower holoblastic forms amongst the Vertebrata in the 

 stricter sense we will consider first Lepidosiren, in which, owing to 

 the large size of the cell-elements, the details of mesoderm formation 

 arc particularly clear and unmistakable. 



The mode of origin of the mesoderm which occurs in Lepidosiren 

 is illustrated by Fig. 35. The section shown in Fig. 35, A is taken 

 from an egg of the same age as that figured on p. 35 (Fig. 21, C) in 

 illustration of the disappearance of the segmentation cavity. Im- 

 mediately below the ectoderm is a mass of rounded blastomeres with 

 intervening chinks remnants of the segmentation cavity : towards 

 the mesial plane the blastomeres are more closely packed together. 

 The small blastomeres in question are clearly distinguished by their 

 finely-grained yolk from the large yolk-cells with their coarsely- 

 grained yolk which form the hulk of the egg. The mass of small 

 blastomeres is destined to give rise laterally to the mesoderin and 

 mesially to the notochord. It must clearly be borne in mind that 

 the mass is composed simply of small blastomeres and that it passes 

 at its outer margin without any break into the ordinary yolk-cells. 



As development goes on, the mass of small elements becomes 

 compacted together (Fig. 35, B), the chinks between the cells disap- 

 pearing. At the same time the boundary between them and the 

 yolk -cells becomes more definite, so as to delimit more clearly the 

 mesoderm rudiment (mes) from the definitive endoderm. 



Fig. 35, C is taken from an egg of the same age but here the 

 mesoderm rudiment has become limited also on its mesial side by a 

 split which marks it oft' from the notochord (N). 



At a somewhat later stage, the mesoderm mass on each side 

 becomes divided into segments by splits, transverse to the axis of the 

 body, which make their appearance at regular intervals from before 

 backwards, but it is to be noted that in Lepidosiren (as in all Verte- 

 brates except Amphioxus) this splitting of the mesoderm is confined 

 to its dorsal portions. There is thus produced along each side of the 

 body a series of incomplete mesoderm segments 1 which pass at 

 their lower or ventral ends into an unsegmented sheet of " lateral " 



1 Such incomplete mesoderm segments as are described above occur in all the 

 typical vertebrates and are known by various names such as mesoblastic somites, 

 protovertebrae, myotomes. These names are in various degrees erroneous or mis- 

 leading. The word somite means a complete body segment and it is not allowable to 

 apply it to a single organ. The name protovertebra dates from the days in which 

 these structures w<-re supposed to be the embryonic vertebrae, which they are now 

 known not to be. Of the three terms mentioned myotome is the least objectionable 

 U it least the greater part of the segmented portions of mesoderm become definite 

 myotomes later on. On the whole however it seems most convenient to retain the ex- 

 pression mesoderm segment, the word segment not being necessarily used in the 

 precisely defined way in which such a purely technical morphological term as 

 " somite " must be used. 



