434 Dr. W. B. Benham on some 



about the alimentary system is presented by the conditions of 

 the ccecum. It is usually stated that these pouches lie in the 

 twenty-sixth segment. From an examination of my material 

 I believe that they in reality belong to the twenty-seventh 

 segment, but, being closely adpressed to the intestine in this 

 segment and passing through the septum in front, so as to lie 

 freely in segment xxvi., it appears at first sight that they 

 belong to this latter segment. In the new species herein con- 

 sidered I carefully examined this point, and find that the 

 ccecum originates from the intestine in the tioenty -seventh seg- 

 ment ; the communication is wide and occupies nearly all this 

 segment. 



In the present worm the caecum is of considerable size ; it 

 rises by a broad base in the twenty-seventh segment, runs 

 forward througli the twenty-sixth, and extends to the anterior 

 boundary of the twenty-fifth segment. Its wall is not entirely 

 smooth, but along its dorsal (inner) and ventral (outer) side 

 is notched, and these notches on the outer side become deeper 

 as the base of the caecum is approached, so as to produce 

 secondary cajca or finger-shaped lobes. Of these one arises 

 in the twenty-sixth segment close to the posterior septum, the 

 rest arise in the twenty-seventh segment. There are about a 

 dozen of these finger-shaped secondary cseca, arranged along 

 the outer side of the base of the cgecum and tending to become 

 smaller as the lower ventral surface of the gut is reached. 

 Some of these secondary ceeca arise as bifurcations of a 

 simple lobe (as in figure 2 c). 



Hitherto only a iQ\N species have been recorded in which 

 more than a single pair of cteca are present, viz. P. Hilgen- 

 dorfii^ F. Sieholdiij P. Gulielmi, P. mandhorensis, and occa- 

 sionally in P. musica^ and in a new species of Beddard's — 

 P. trityphJa — from the Sandwich Islands. Probably in all 

 these cases the mode of origin of these additional ca3ca is 

 the same, as they are described as lying one above the other, 

 to the number of three to six, according to the species ; but in 

 none are they so numerous as in the present species, hence 

 the specific name " digitata^ 



The vascular system presents the following features : — 



There are two pairs of latero-intestinal hearts in segments 

 xii. and xiii., each being connected dorsally with the large 

 dorsal trunk and the much smaller supra-intestinal vessel. 

 In segment xi. there is an " intestinal heart " connected 

 only with the supra-intestinal vessel. The dorsal vessel con- 

 tinues forward as a narrow tube above the gizzard, giving off 

 a pair of vessels just anterior to the septum x./xi. ; a second 

 pair in front of tlic hinder margin of the gizzard ; another 





