July 31, 1884] 



NA TURE 



329 



cavity into two anterior chambers and one posterior ; and lastly 

 a transverse septum shuts oft" the space within the epistome and 

 tentacles from the rest of the body-cavity. The nephridia open 

 into the posterior chamber of the body-cavity on the sides of the 

 lateral mesentery. At no stage, either in the embryo or the 

 adult, is any trace to be found of segmentation. 



The parallelism between Phoronis and Brachiopoda is full and 

 clear. An ectodermal post-oral nerve-ring exists in both. The 

 body-cavity of the prae-oral lobe is in both separated from the 

 rest of the body-cavity by a septum. The tentacles are arranged 

 and developed similarly in both. In both the nephridia have 

 the same relations and the alimentary canal is divided into the 

 same four parts. And in both the prce-oral lobe of the larva is 

 represented in the adult by an epistome. The Polyzoa, though 

 immensely simplified in structure, seem undoubtedly to be built 

 upon the same plan ; and Caldwell considers it probable that 

 the epistome of Endoproct and Hippocrepian Polyzoa and the so- 

 called foot in Rhabdopleura represent, like the epistome of 

 Phoronis and Brachiopods, the pine-oral lobe. 



Mr. Caldwell closes the abstract of his yet unpublished paper 

 with a remark upon the affinities of the Gephyrea. We know 

 nothing to show that Sipunculus and Phascolosoma are not 

 referable to the same type of structure as Phoronis, Brachiopoda, 

 and Polyzoa. 



But as regards the types from which all these mutually-con- 



Diagram of body plan of Brachiopod, Polyzoon, and Phormiis (after Cald- 

 well).—)//, mouth ; a, anus ; s, septum ; n. nervous system ; st, 

 stomach ; p, second stomach ; cp, epistome ; v, tentacle of ventral 

 series ; d, tentacle of dorsal series ; v V valves of Brachiopod skull, 



nected forms sprang, we know little or nothing, and we look in 

 vain for an unsegmented worm which shall show clear affinity 

 with them. 



Van Bemmelen, in a recent paper {Jcnaische Zei/schrift, 18S3) 

 has compared at great length the Brachiopods with Sagitta, and 

 arrives at the conclusion that the two types show such intimate 

 agreement that they must be looked upon as very closely related. 

 In the first place Dr. van Bemmelen recounts the histological 

 resemblance between Sagitta and Brachiopods ; and if he 

 ascribes more weight to these than his readers may be inclined 

 to do, he is not without weightier considerations in support of 

 them. In both groups connective tissue is conspicuously scanty ; 

 in both a homogeneous intercellular substance or mesenchyme- 

 layer is abundant. The epithelial layers are extremely simple 

 and alike in both ; the muscles in both are " built on an epithelial 

 type ; " the histological characters of the nervous system are 

 identical in both. The chitinous hairs developed in ectodermal 

 follicles on the mantle of Brachiopods are not without analogues 

 in the Chaetognatha. The three metameres of the larval 

 Brachiopod are compared with the divisions of the adult Sagitta ; 

 the four genital glands of the former (Testicardines) are identified 

 with the ovaries and testes of the latter. The gastro- and ileo- 

 parietal bands of the Brachiopod are made homologous with the 

 transverse septa of Sagitta, and the hood of Sagitta with the 

 arms of the Brachiopod. It is obvious that the above characters, 

 many of which are mentioned in the Hertwig's " Ccelomtheorie,'' 



are of very unequal value ; and some are even wrong, for th 

 gastro- and ileo-parietal bands of a Brachiopod are parallel with 

 the gut, and in no way comparable to the transverse septa of 

 Sagitta. But, on the other hand, there are other suggestive 

 points of resemblance, and, though further developmental 

 evidence in the case of Sagitta is sorely needed, I think that its 

 possible affinity to the Phoronis type cannot be altogether passed 

 over. Not only is the development of the mesoblast and body- 

 cavity strikingly similar, but the dorsal and ventral mesenteries 

 at first present in Phoronis agree with those of Sagitta, and the 

 septum dividing off a part of the body-cavity within the head 

 seems the same in both. Nothing in the nervous system offers 

 great difficulty, and the relations of the hood in Sagitta seem not 

 discordant with those of a lophophore. If we approximate the 

 anus and mouth dorsally in Sagitta, the "olfactory organ" will 

 assume the position of the two sense-organs of Phoronis and 

 Rhabdopleura. The lateral mesenteries of Phoronis and Brachio 

 pods arise late and secondarily, as does the transverse septum 

 of the trunk in Sagitta. The anterior and posterior generative 

 masses, arising first together, are no sign of true segmentation, 

 and our embryological knowledge of the nephridia of Sagitta is 

 too slight to permit us to make much use of them as arguments 

 on either side. 



If all this is true (and I am far from insisting upon it), it 

 means that Sagitta (though extremely modified for a pelagic life) 

 is akin to the unflexed, unsegmented worm, which, as it acquired 

 a dorsal flexure and a more complex lophophore, gave rise to 

 the proximate ancestors of the Phoronis type. 



Sedgwick's theory of segmentation : A. Ideal a 

 mals ; B. Invertebrate. — M, primitive- mouth 



mdia , 11. iiivciLtuidtt — ..i. juimuive muuui , ///. moutn ; ti, anus; M, 

 middle portion of primitive mouth or blastopore closed up; K, nervous 

 system ; p, pouch of gut ; ms, mesoblastic somite ; k, nephridium ; o,. 

 external pore ; "- 



And if we admit this even for a moment, it becomes worth 

 while to consider the possibility of a distant Molluscan connec- 

 tion with the same line ; for, possessing a trochosphere larva, a 

 single pair of nephridia, and a nervous system with no trace of 

 genuine segmentation, they are so far in agreement with our 

 type. I cannot see that Caldwell's discoveries necessarily 

 invalidate Lankester's old comparison of the Lamellibranch gills 

 (and labial palps) with a lophophore ; and even Lankester him- 

 self, in spite of his opinion already quoted, that, owing to Cald- 

 well's research, Polyzoa, &c, must now be removed from the 

 Molluscan phylum, yet still admits (/oc. nil. p. 688) that "it is 

 very probable that the labial tentacles and gill-plates are modi- 

 fications of a double horseshoe-shaped area of ciliated filamentous 

 processes, which existed in ancestral Mollusks much as in 

 Phoronis and the Polyzoa, and is to be compared with the con- 

 tinuous prae- and post-oral ciliated band of the Echinid larva 

 Pluteus, and of Tornaria ; " and Langerhans' close comparison 

 between the nervous systems of Sagitta and a Mollusk may be- 

 worth more consideration. The molluscan foot may, after all, 

 be an epistome, as Lankester formerly said, and the "osphradia" 

 of the Mollusk may turn out homologous with the sense-organs 

 of Phoronis and Rhabdopleura. But the extreme modifications 

 that the Molluscan type has undergone — the reduction of the 

 body-cavity, the development of the foot, the various flexures, and 

 so forth — leave any connection that we may trace with it and 

 our Sagitta type at best a distant one ; if such exists, a distant 

 relationship will be again traceable between Mollusks and 



