LEWIS. — POLYCH^TE ANNELIDS. 247 



The place of emergence of this process is not constant, but usually it 

 seems to arise from the cell at a point between the centrosome and the 

 nucleus. Figures 33, 33% 34 (Plate 5) show the relation of cell-process 

 to nucleus and centrosome in several cases. 



The most important facts regarding the giant cells of the nervous 

 system of these two worms may be summarized as follows : — 



1) The giant cells give rise to Leydig's fibres. 



2) They are present in very limited numbers throughout the nerve 

 cord ; they show no regularity of arrangement, and no constancy in num- 

 ber in the different segments. 



3) They are all characterized by the possession of a nucleus, eccen- 

 tric in position, and by the presence of a centrosome and sphere. 



III. PERIPHERAL NERVOUS SYSTEM. 



1. Introduction. 



This part of the paper, dealing with the peripheral nervous system of 

 the two Maldanida^, Axiothea torquata and Clymene producta, was pre- 

 sented in substance before the American Morphological Society in Decem- 

 ber, 1896, and was afterwards summarized in Science, Vol. V. No. 115, 

 p. 428. 



Under the term " epidermal sense organs " I include certain organs 

 found abundantly on all the segments of the body in both worms men- 

 tioned. Similar organs have been referred to in the case of the head and 

 proboscis of certain marine annelids. One writer only, so far as I know, 

 has mentioned such organs as being found on other parts of the body of 

 any marine annelid. Eisig ('87, p. 547) has described similar organs 

 from both the abdominal and thoracic segments of two of the Capitellidse. 

 No one has given any account of the plan of distribution of such organs, 

 nor has their structure, as determined by the newer nerve methods, been 

 described. I have examined the removed cuticula of four marine worms 

 representing three different families, and have found evidences of such 

 organs in all parts of the integument of the worm ; I conclude, therefore, 

 that, if not universally present on the body of polycha^tes, they are at 

 least very common in marine worms. 



2. Methods Employed. 



In Clymene and Axiothea, three principal methods of study were 

 employed, the results obtained by one being confirmed and extended by 

 the others. They were as follows : (1) maceration, by which the cuticula 



