268 PROCEEDINGS OF THE ACADEMY OF [April, 



(figs. 69 and 70). Little change takes place in the first five parapodia, 

 but with the sixth the whole parapodium begins to diminish in size, 

 the neurocirrus especially dwindling until by the ninth the entire style 

 has disappeared and the base is represented only by the usual opake, 

 somewhat swollen, whitish area ventral to the base of the foot, which 

 becomes smaller but continues to the caudal end. The postsetal lip 

 becomes smaller very gradually, but remains to the middle segments 

 as a small conical process. Except that it becomes more slender and 

 assumes the proportions of a gill filament, the notocirrus undergoes 

 little change (fig. 71). 



Branchia? appear on both sides of somite V (fig. 70) as a single 

 filament as long, but much more slender than the postsetal cirrus, 

 arising from the notopodial base in common with and on the dorsal 

 side of the notocirrus. On succeeding segments the filament becomes 

 long and on IX a second appears; additional ones then appearing (sym- 

 metrically except as mentioned) up to the number of nine, as follows: 

 three on XIII (XIV on right), four on XVI, five on XIX, six on XX, 

 seven on XXI, eight on XXVII and nine at about XXXV. The last 

 number (fig. 71) continues to at least L. The largest gills on pieces 

 from the middle of the body, the segment numbers being undetermined, 

 bear as many as twelve filaments and the most caudal segments 

 represented bear unifilar gills. Until they possess upward of four 

 filaments, the gills scarcely exhibit a main stem or pinniform structure 

 which is always obvious on the more complex gills. The main stem 

 curves rather sharply dorsad, tapering, and bearing along its lateral 

 side the filaments, which diminish in size to the last. On anterior 

 segments the filaments are shorter than the notocirrus, but farther 

 back they are longer. New ones appear to be added from the growing- 

 point at the tip of the stem. Branch blood-vessels from the main 

 trunk enter all of the filaments. 



Neuropodial acicula, which on anterior parapodia are not very 

 clearly distinguished from the setse, are from three to five in a row, 

 rather stout, tapered gently to near the end and then abruptly into 

 slender, very acute projecting tips, appearing at the bases of the 

 dorsal seta?. Notopodial acicula are very long and slender fibers 

 which enter the base of the notopodium and continue far into the 

 cirrus (fig. 71). Perhaps they would be more correctly described as 

 buried setse. 



Except the large ventral crochets, all seta? are colorless or nearly so. 

 All segments bear a small tuft of slender, acute, capillary seta? dorsal 

 to the acicula, among the bases of which are a very few much more 



