112 DR. E. B. WILSON ON THE DEVELOPMENT OF RENILLA. 
§.10. Development of the mesenterial filaments. 
The mesenterial filaments are visible as soon as the larva becomes sufficiently 
transparent as dark granular thickenings on the edges of the septa at their upper 
portions where they join the cesophagus. They may be seen while the larva is still 
swimming, but their arrangement can be made out only after the larva has attached 
itself and the body has begun to elongate. It is then apparent that they vary in 
length and have a definite disposition. Those of the dorsal septa are very short 
indeed (fig. 177), or in some cases may not be visible at all when the others are well 
developed. They appear as knob-like prolongations of the lip of the cesophagus 
attached to edges of the dorsal septa. The dorso-lateral filaments (d,/.f.) are much 
longer, extending along the edges of the septa nearly to the buds (p!.) which have 
now appeared at the point where the dorsal and dorso-lateral septa unite. The ventro- 
lateral filaments (v./.,f:) are still longer and extend down to the level of the buds or 
beyond them. The ventral filaments, finally, are very short, being intermediate in 
length between the dorsal and dorso-lateral filaments. 
This grouping of the filaments is quite constant and exists at a very early stage. 
It is extremely difficult to determine whether these varying lengths represent the 
actual succession of the filaments since the latter are in their early stages closely 
contracted together and their arrangement cannot be made out. This grouping 
persists for a long time and the dorsal filaments remain permanently shorter than the 
others, and of different structure as KOLLIKeER has observed. (The dorsal filaments 
are in many cases longer than the others, but this is, in Renilla at least, only 
apparent, and is due to the fact that they never become convoluted like those of 
the lateral and ventral septa.) All of the filaments except the dorsal pair increase 
rapidly in length and very soon become folded back and forth and variously convoluted 
(see figs, 183, 205). This is a result of the circumstance that the filaments increase 
in length much more rapidly than the septa which bear them, and they are necessarily 
therefore thrown into folds or “ gathers.” 
The dorsal filaments grow backwards very slowly and are never thrown into 
transverse folds (see figs. 183, 204). They are less opaque than the other filaments, 
with a darker central line, and are of much less diameter than the others. These 
differences are permanent and persist in the adult. The dorsal filaments always 
remain in connexion with the cesophagus, and appear like long narrow prolongations 
of the latter down upon the edges of the septa. The other filaments, though at first 
extending quite up to the cesophagus, soon become more or less widely separated from 
the cesophagus, fading insensibly away a short distance below the lips of the latter. 
In transverse sections the filaments appear as simple thickenings of the entoderm 
at the edges of the septa, which differ in appearance from the remaining entoderm of 
the septa only in being more granular. The supporting lamella may be traced out 
nearly to the middle of the thickening where it fades away and disappears, 
