XII] EMBRYOLOGY 237 



ectodermal. If this be so, the alimentary canal is entirely of 

 ectodermal origin, and no true endoderm exists in the Odonata. 



Students interested in these rival theories will find them fully 

 discussed in Korschelt and Heider [81] to whom they are referred. 

 Supporters of Gaskell's theory will, no doubt, see in Heymons' 

 statements some further evidence for their views. 



Formation of the Segments and Appendages (fig. 109 D G). 



After the first appearance of the cephalic lobes, and the 

 completion of the sinking-in of the ventral plate, the embryo 

 gradually elongates at the end opposite to the future head, the 

 extreme end becoming curved over as in D. This curvature 

 remains for the rest of embryonic life. On the curved end a 

 slight enlargement marks the anal plate, on which, soon afterwards, 

 the proctodaeal invagination begins to form. The body of the 

 embryo very soon begins to shew signs of segmentation, by the 

 formation of slight transverse constrictions, indicating the limits 

 of the future body-segments. The constrictions only shew up on 

 the ventral surface of the embryo, which tends to become more 

 convex. On either side of the mid-ventral line, the convexity 

 indicating a future segment very soon begins to take the form of 

 a pair of latero-ventral swellings the first rudiments of the 

 appendages. The first distinct swellings to be seen are three pairs 

 at about the middle of the ventral surface of the embryo. These 

 represent the rudiments of the future thoracic legs (E, l : , 1 2 , 1 3 ). 

 Next to these, nearer the head, three very similar swellings can 

 be made out, destined to become the appendages of the mouth 

 (E, md, mx 1} mx 2 ). At the same time, the cephalic lobes alter in 

 shape. From their more ventral part a pair of swellings arises 

 similar to that mentioned above, but destined to form the 

 antennae. Just in front of this, a median unpaired swelling 

 arises, forming the front of the head. The stomodaeal invagination 

 takes place just posterior to this. 



The appendages gradually elongate, but at unequal rates, the 

 most posterior pair (third legs) soon becoming the longest, while 

 the others decrease from behind forwards in order, up to the 

 mandibles (F). The antennae, however, soon become longer than 



