Evolution of Animals 423 



elongated tube that was gradually separated from the ali- 

 mentary canal or blastocoel, through strains exerted on its 

 walls by the extensor and retractor muscles of the proboscis, 

 and by sliding action of the proboscis itself. Favoring this 

 view — apart from confirmatory details furnished by turbel- 

 larians — is the occurrence in Drepanophorus of lateral lobes 

 that correspond in position and verticality with coeca of the 

 alimentary tract. The structure of the sheath equally sug- 

 gests its embryonic origin, for this consists of an investing 

 cell layer similar to that lining the alimentary canal, that is 

 surrounded by an inner longitudinal and outer circular mus- 

 cular wall. In most genera the sheath contains loose irregular 

 semistellate cells, but Hubrecht draws attention to the fact 

 that in Cerehratulus its posterior end, that is not now occupied 

 by the proboscis, becomes filled mth soft continuous nucleated 

 cellular tissue, much resembling the central cells of the verte- 

 brate notochord. This is a suggestive partial proof that, 

 where the proboscis has come to fill its sheath imperfectly, 

 the latter instead of being obliterated has so fully assumed 

 another function — that of strengthening in embryonic and 

 later life — that it increases its cellular continuity, instead of 

 becoming absorbed. 



As Hubrecht, Burger {132), and others have also empha- 

 sized, the sheath extends lengthwise anteriorly to "the level 

 of the nervous commissure, through which the proboscis passes." 

 Further in its disposition amongst typical vertebrates "the 

 anterior end of the notochord terminates immediately behind 

 the attachment of the pharyngeal membrane" {ISJ^.: 436). 

 This is typically the line of insertion of the proboscis into the 

 anterior end of the sheath in nemerteans, so that in both details 

 given the two structures accurately agree. 



Hubrecht viewed the proboscis in its entirety as "homolo- 

 gous Tvnth the rudimentary organ which is found in the whole 

 series of the vertebrates without exception — the hypophysis 

 cerebri" (pituitary body). This brilliant explanation might 

 have obtained rapid and wide acceptance had its author some- 

 what limited or modified its scope, and at the same time pushed 

 his comparisons further. 



That the proboscis for some Nemertinea and the hypophysis 

 for vertebrates are both primarily developments from the 

 roof of the oral sinus are now accepted facts. Thus Hertwig 

 says regarding the latter {pp. cit. 436): "It first arises at a 

 very early stage, when the oral sinus has just arisen, and is 

 still separated from the cavity of the head gut by the phar- 

 yngeal membrane. It is therefore a product of the outer 



