424 Causes and Course of Organic Evolution 



germinal layer and not a growth from the cavity of the head 

 gut as earlier had been maintained." Also "it remains a long 

 time in connection with the oral cavity by a narrow duct." 



The proboscis in the fresh-water and land forms as well as 

 in such nemertean genera as Amphiporus results from doisal 

 uppitting and backward growth of the pharyngeal area, in 

 the front part of the gullet, the whole extending backward 

 above the alimentary^ canal and beneath the dorsal nerve cord. 

 Capacity for protrusion of the eversible anterior part gives 

 rise to the proboscis. Now it is in the genera indicated above 

 that primary and accessory rods or stilets — the importance 

 of which phylogenetically we hope soon to emphasize — orig- 

 inate; that the proboscis is often richly loaded mtli nemato- 

 cysts; and in which the mouth is in front of the brain. 



We would regard the shifting of the proboscis upward ante- 

 riorly, and its coming to open by a distinct orifice in line TN-ith 

 the sheath but in front of the mouth, as a derived arrangement 

 evolved because it then did not interfere mth the direct use 

 of the mouth as an absorptive structure. Such genera theie- 

 fore as Eupolia and Valencinia are modified in this respect 

 from a more primitive stock. 



In order to facilitate further description we would now draw 

 attention to the accompanying diagram, of some nemerteans 

 and simpler vertebrates, as favoring Hubrecht's view and ex- 

 tending it in various directions (Plate Fig. 15). 



In emphasizing the direct homology of the hypophysis of 

 vertebrates with the proboscis of nemerteans, Hubrecht does 

 not seem to have considered the possibility hereditarily of only 

 a part of it being thus utilized in transition from nemerteans 

 to primitive cyclostome vertebrates, and of a remainder being 

 utilized in other ways. We may best ascertain if such is likely 

 after examination of the proboscis when at rest, and what 

 might result if, by gradual filling up of the sheath as we see 

 proceeding in Cerebratulus, it thus becomes converted sec- 

 ondarily into a dorsal sensitive and strengthening oigan, thus 

 causing the entire proboscis to be extruded and again united 

 wdth the oral cavity. 



Plate Fig. 15, f is a diagram of the proboscis extruded in 

 front of the now solid notochord, and separating into the pos- 

 terior glandular portion that is pushing upward from the roof of 

 the mouth, and underneath the infundibulum. This trans- 

 formation seems a necessary step in evolution from the nemer- 

 teans to a group somewhat distantly related to the cyclostomes 

 and which in the figure we have termed a paracyclostome type. 

 The anterior and originally inverted part of the proboscis 



