OCTOHEK 8, 1896] 



NA TURE 



563 



E xiretory Organs and 

 Calami, Cavities. 



Coxal gland 



1st head cavit)f, pKBoral... 



2nd head cavity. Cavity 

 of prosomatic segments 



Cavities to each meso- 

 somatic segment 



Ifearl and Vascular Sysleiii. 

 Dorsal heart 



Longitudinal venous sin- 

 uses 



Lacunar blood spaces of 

 appendages 



Pituitary gland. 



1st head cavity, pr.tnral. 



2nd head cavity, mandibular. 



Cavities of hyoid and branchial 

 segments. 



Column of fatty tissue dorsal to 

 spinal c(jrd. 



1 Icart and ventral aorta. 



Lacunar blood spaces in velar 

 and branchial appendages. 



J'lie Possible Meaning; of the Notochord. 



Although we can .say that every structure and organ in the 

 prosomatic and mesosomatic regions of Limulus, &c. , is to be 

 found in the head region of Ammocretes, we cannot assert the 

 reverse proposition, that every organ in the head region of 

 Ammocretes is to be found in Limulus, &c., for we find a not- 

 able exception in the case of the notochord, a structure which is 

 far excellence a vertebrate structure, and has in consequence 

 given the current name to the group. Such a structure is clearly 

 not to be found in Limulus and its allies ; it has evidently arisen 

 in connection with the formation of the vertebrate alimentary 

 canal from the oral and branchial chambers, and it evidently at 

 one time possessed a functional significance, for the lower we 

 descend in the vertebrate scale the more conspicuous it becomes. 



Unfortunately we know nothing of the condition of the noto- 

 chord in the early extinct fishes, sr) that we are reduced to the 

 embryological method of inquiry in our endeavours to find out 

 the meaning of this organ. This method appears to point to the 

 origin of the notochord from a tube connected with the alimen- 

 tary canal, originally therefore an accessory digestive tube ; the 

 reasons why such a view has been put forward are, first, the 

 origin of the notochord from hypoblast ; secondly, the evidence 

 that it is to a certain extent tubular ; and thirdly, that it is an 

 unsegmented tube extending from the oral to the anal regions of 

 the body. Another argument, to my mind stronger than any 

 other, is based on the principle that nature repeats herself, and 

 if, therefore, we find the same proliferation of cells in the same 

 place forming a series of solid notochordal rods, we may fairly 

 argue that we are observing a series of repetitiims of the same 

 process for the same object. Now the formation of the head 

 region of Petromyzon shows that at first a median proliferation 

 of hypoblastic cells occurs to form the notochord, which then 

 se|)arates off from the hypoblast ; later on a similar proliferation 

 takes place to form the subnotochordal rod, which similarly 

 separates off from the hypoblast ; later still, at the time of 

 transformation, a third median proliferation of the cells of the 

 hypoblast takes place, to form a solid rod of cells. This solid 

 rod then commences to hollow out at the end nearest the 

 intestine, and the hollowing out process extends gradually to 

 the oral end, until a hollow tube is formed connecting the mouth 

 with the intestine. In this way the new gut of the adult 

 Petromyzon is formed from a solid median rod of cells closely 

 resendiling in its formation the original notochord. 



I put \\ forward therefore as a suggestion, that in the ancient 

 times when the Merostomata were lords of creation and the com- 

 petition was keen among these ancient arthropod forms, in 

 which the nervous system was so arranged that increase of brain 

 substance tended more and more to compress the food channel, 

 and therefore to compel to the suction of liquid food instead of the 

 mastication of solid, accessory digestive apparatuses were formed, 

 partly in connection with the formation of the oral respiratory 

 chambers, and partly by means of the formation of the notochord. 

 Of these accessory methods of digestion the former became per- 

 manent, while the latter becoming filled up with the peculiar 

 notochord.il tissue became a supporting structure, still showing 

 by its unsegmented character its original function. That a tube 

 formed from the external surface either as notochord or as the 

 respiratory portion of the alimentary canal in Ammocretes should 

 be capable of acting as a digestive tube is clear from the 

 researches of .Miss Alcock {Proc. Camb. Phil. Soc, vol. vii., 

 1891), for she has shown that the .secretion of the skin of 

 Ammocretes easily digests fibrin in the presence of acid. Such 



NO. 1406, VOL 5 ].] 



a secretion, like the similar secretion of the carapace of Daphnia 

 and other crustaceans, -was originally for the purpose of keeping 

 the skin clean. 



The evidence which I have put before you is in agreement 

 with the conclusion that the fore gut of the vertebrate arose 

 gradually from a chamber formed by the lamellar branchial 

 append.^ges, which functioned also as a digestive chamber. By 

 the growth of the lower lip, or metastoma, and the modification 

 of the basal portion of the last locomotor appendage, which 

 basal part was inside the lower lip, into a valvular arrangement 

 like the velum, the animal was able to close the opening into 

 the respiratory chamber and feed as blood sucker in the way of 

 the rest of its kind, or when living food was scarce, keep itself 

 alive by the organic material taken into its respiratory chamber 

 with the muddy water in which it lived. 



The Possible Formation of the 1 'ertehrate Spinal Region. 



It remains to briefly indicate the evidence as to the formation 

 of the rest of the alimentary canal and the spinal region of the 

 body. 



The problems connected with the formation of this region are 

 of a different nature from those already considered in connection 

 with the cranial region. 



In the cranial region the variation that has taken place within 

 the vertebrate group and in course of the formation of the verte- 

 brate is, on the whole, of the nature called by Bateson sub- 

 stantive, i.e. increase or suppression of parts, while throughout 

 the parts remain constant in their relations to each other. Il 

 matters not whether it is frog, fish, bird, or mammal we are con- 

 sidering ; we always find the same cranial nerve supplying the 

 same segments. When we consider the spinal cord and its 

 immediatejuncticm with the cranial region, this is no longer .so ; 

 here we find a repetition of .similar segments, with great variation 

 in the amount of that repetition ; here we find the characteristic 

 feature is meristic variation rather than substantive, and so inde- 

 termined is the vertebrate in this respect that even now the same 

 species of animal varies in the number of its segments and in the 

 arrangement of its nerves. In this part of the vertebrate body 

 this repetition is seen not only in the central nervous system and 

 its nerves, but also in the excretory organs, so that- embryology 

 teaches us that the vertebrate body has grown in length by a 

 series of repetitions of similar segments formed between the head 

 end and the tail end ; such lengthening by repetition of segments 

 has been accompanied by the elongation of the unsegmented 

 gut, of the unsegmented notochord, and of the unsegmented 

 neural canal. 



To put it shortly, all the evidence points to and confirms the 

 view so strongly urged by Gegenbauer, that the head region is 

 the oldest part and the spinal region an afterthought, that the 

 attempt so often made to find verteljrce and spinal nerves in the 

 cranial region is an attempt to put the cart in front of the horse 

 — to obtain youth from old age. We may, it seems to me, 

 fairly argue from the sequence of events in the embryology of 

 vertebrates that the primitive vertebrate form was chiefly com- 

 posed of the head region, and that between the head and the 

 tail was a short body region. In other words, the respiratory 

 chamber and the cloacal region were originally close together, 

 just as would be the case in Limulus if the branchial appendages 

 formed a closed chamber. According, then, to my view, there 

 would be no difficulty in the respiratory chamber opening 

 originally into the cloacal region, i.e. the same cloacal region 

 into which the neurenteric canal already opened. The short 

 junction tube thus formed would naturally elongate with the 

 elongation of the body, and, as it originally was part of the 

 respiratory chamber, it equally naturally is innervated by the 

 vagus nerve. This, then, is the explanation of that most extra- 

 ordinary fact, viz. that a nerve essentially branchial should 

 innervate the whole of the intestine except the cloacal region. 

 Whether this is the true explanation of the formation of the 

 mid-gut of the vertebrate cannot be tested directly, but certain 

 corollaries ought to follow : we ought to find, on the ground 

 that the sequence of the phylogenetic history is repeated in the 

 embryo, that ( i ) the growth in length of the embryo takes place 

 between the cranial and sacral regions by the addition of new 

 segments from the cranial end ; (2) the formation of the fore-gut 

 and hind-gut ought to be completed while the mid-gut is still an 

 undifferentiated mass of yolk cells ; (3) the cloacal region ought 

 to be innervated from the sacral nerves, while the stomach, mid- 

 gut and its appendages, liver and pancreas, ought to be inner- 

 vat'jd from the vagu^. 



