SKELETON AND BODY WALL 447 



The constitution of this body wall shows up very clearly when examined by polarized 

 light. I made a microtome preparation of the right half of Doloria and, after removing 

 the wax, and transferring the specimen to glycerine, picked out the gut and gonads 

 with fine needles and washed out the body cavity. Such a specimen, when examined 

 between crossed Nicols, shows either the dorsal longitudinal muscles brilliantly illu- 

 minated and the circular muscles as dark streaks across them, or the reverse, depending 

 on the relation between the direction of their myofibrils and the axes of the Nicols. 

 A more satisfactory method of study is to place a thin selenite layer underneath the 

 lower Nicol. In this case one set of muscles will appear brightly coloured and the other 

 set also bright, but in a different colour. By rotating the stage the colours can be 

 reversed. In this way the circular muscles can be followed very accurately to their point 

 of fusion with the ectoderm. They do not appear to be so markedly doubly refractive 

 as the dorsal longitudinal muscles. 



The extremely close association of the circular muscles with the cuticle, and the fact 

 that it is difficult to decide exactly where the muscle emerges from the ectoderm, sug- 

 gests that these muscles are of ectodermal origin. Further in support of this, is their 

 staining reaction and histological detail. The dorsal longitudinal muscles stain on the 

 whole a deep purplish red, and show all the detail of typical striped muscle. The circular 

 muscles, on the other hand, stain a brighter red and exhibit merely an alternation of 

 staining and non-staining zones which merge into each other. Krause's membrane is 

 entirely absent. Now these are just the characteristics of the early stages of those dorso- 

 ventral muscles of Chirocephaliis which I was able to show (1926 b, p. 406) were derived 

 from the ectoderm. They also exhibited this "moniliform" staining and did not show 

 Krause's membrane during their early stages. 



Further, I showed that in still earlier stages of these muscles the myofibrils were 

 continuous and stained uniformly throughout their length. Just behind and above the 

 attachment of the adductor muscle of Doloria, sections tangential to the surface show 

 bands of such continuous myofibrils running in various directions, and these appear 

 to lie in the ectoderm cells. I consider that these are ectodermal muscle cells which 

 are even less "mature" than the circular muscles. 



The chief interest in the dorsal body wall of Doloria lies in the fact that, in its con- 

 stitution, it is Annelidan. In the Annelid the body wall consists of a thin cuticle, the 

 ectoderm, the circular muscles and the longitudinal muscles occurring in this order 

 from outside inwards, and the circular muscles of Criodrihis are ectodermal according 

 to Staff (1910). In Doloria there is the same arrangement. I do not mean to imply 

 that the circular muscles of Doloria are homologous with those of Annelids. From my 

 work on Chirocephaliis I believe, although I did not stress this point in the paper, that 

 the Annelid circular muscles became transverse and dorso-ventral muscles when the 

 cuticle hardened and took over the function of supporting the musculature, and the 

 Annelid became thus transformed into an Arthropod. The ancestral Crustacean pro- 

 bably had this muscle constitution. But in Doloria the whole body has become enclosed 

 in a protective shell. All the limbs, the mouth-parts and the eyes are supported by the 



