62 THE APODID^ parti 



of the septa (see Fig. 15, B), and are probably 

 longitudinal muscles. The order in which these 

 latter groups of muscles occur is significant of their 

 origin, as shown in the diagram (Fig. 16). We find 

 that those which arise most dorsally run the furthest 

 into the limbs, this rule being regularly observed. 

 This order is what we should expect if we assume 

 that these are parts of the longitudinal musculature 

 which ran outwards into the parapodium, the dorsal 

 edge of which was then gradually lengthened, and 

 the whole turned round the body in the transverse 

 plane towards the ventral middle line, as shown in 

 Fig. 16. The dorsal muscle bands will naturally be 

 the most lengthened and reach the furthest down 

 into the limb ventrally. In the case of longitudinal 

 muscle bands running outwards into the parapodia, 

 but traversing them from the anterior to the posterior 

 wall, the same rule would hold and the bands would 

 cross one another, as shown in Fig. 15, C} 



The musculature running into the ventral part of 

 the limb or the ventral parapodium is more easy to 

 separate into its elements (see Fig. 14). The longi- 

 tudinal muscles come direct from the ventral muscle 

 bands, and run sloping backwards, as shown in Fig. 12, 

 so that the ventral parapodia or gnathobases which 

 slope away downwards and backwards may be used 

 for pushing food forwards in the middle line. The 



"i We have, however, only once found such muscle bands, and have 

 since repeatedly looked for them in vain. It is not unlikely that our 

 observations relating to them were incorrect, being founded on a series 

 of sections, through which the individual muscle bands had to be 

 followed, 



