NO. 4 HARTMAN : POLYCHAETOUS ANNELIDS 395 



Contrary to the view expressed above, regarding the function of the 

 hood (that is, that it probablj^ protects the teeth and fang as the newly 

 developed hook emerges from the hypodermis of the parapodium and that 

 it is thereafter more or less completely shed), it now seems more likely 

 that the functioning may be somewhat as follows. The hood (when 

 present) probably persists normally throughout the existence of the hook. 

 It is composed of a resilient membrane in one piece with distal aperture 

 and a long slit along the distal end of the hooked side (pi. 43, figs. 1,2). 

 The distal oval aperture is more or less delicately serrated at its margin 

 and is capable of being somewhat compressed in line with the cleft, when 

 the hooks of a series are withdrawn into the parapodium so that the beak 

 is covered. On the other hand, when the parapodium is extended and the 

 uncinal torus is pushed outward, the separate hooks are pushed apart, 

 allowing the tension on the hood to be released and the distal aperture to 

 be spread apart, exposing fang and crest. The hooks function, perhaps, 

 to gain traction as the individual moves up and down in the tube or bur- 

 row. When the parapodium is retracted, the protecting hood is drawn 

 firmly about the beak. Foreign particles, such as silt and fine sand grains, 

 are seldom seen in the hood ; it may be assumed that another function of 

 the hood, by reason of its flexibility and structure, is to keep the parts of 

 the hook free from debris. 



Only one species has been found in which a hood is totally lacking, 

 both in immature, embedded hooks and in fully formed functioning hooks 

 (pi. 57, fig. 4), but this species, Protocapitella simplex (see below), is 

 unusual in other respects also. 



Fundamentally, the structure of the hooks is as stated by Eisig. How- 

 ever, no instance has been found, among the hooded hooks examined, in 

 which there are actually several rows of teeth, with as many as 7 in each ; 

 in a few instances there is a tendency for the teeth in a single row to pile 

 on one another (pi. 52, fig. 1) ; rarely they are in several rows (pi. 58, 

 fig. 3 ) . Most commonly they are in a single transverse row that may be 

 slightly arched or crescentic. 



When the hooks are examined for their true 3-dimensional relations, 

 great care must be used in properly orienting them since the usual posi- 

 tion assumed when a series is dissected away from the body is for the 

 individual hooks to lie in partial three-fourths view. This is so not only 

 because they are connected with one another by common membranes, but 

 also because the shaft and beak are usually so constructed that lateral and 

 frontal views are seldom seen. Furthermore, the high magnifications 



