July 19, 1883] 



NATURE 



2 8 5 



factory organ. It is placed on the dorsal edge of the anterior 

 end of the branchial sac, behind the circle of tentacles, and 

 usually in a distinct " peritubercular " area, a diverticulum from 

 the prebranchial zone formed by a bending posteriorly of the 

 dorsal ends of the peripharyngeal bands. 



The dorsal tubercle is, in the simplest form known, a funnel- 

 shaped depression having its wider circular open end separated from 

 the prebranchial zone in front of the branchial sac by a raised edge 

 orlip, while its opposite narrower end is continued into a fine canal 

 running dorsally and posteriorly. This simple condition is found 

 in Molgula pedwtculata ; iziEugyrakerguelerumis the aperture is 

 still wide, although its edge is square in place of being circular. 

 In other simple Ascidians the anterior half of the edge has 

 been apparently pushed backwards, so as to become invaginated 

 and closely applied to the posterior half, thus reducing the 

 circular aperture to a crescentic or semicircular slit. This con- 

 dition is found in Corella parallelogram ma. In most other 

 forms more or less complication is produced by the ends of the 

 slit, or "horns" as they may be called, being prolonged, often to 

 a very great extent, and coiled in various directions, sometimes 

 producing beautifully regular and closely placed spirals. The 

 patterns produced by this curving of the horns are very numerous 

 and often complicated, but their value in classification is slight, 

 since they differ sometimes to a considerable extent in individuals 

 of the same species, and on the other hand are sometimes very 

 similar in members of different genera or even families. 



This variously-shaped organ is histologically merely a depres- 

 sion in the connective tissue of the mantle, lined by epithelium 

 continuous with the squamous epithelium covering the pre- 

 branchial zone, but modified upon the edges of the slit into 

 cubical or columnar ciliated cells. Since the time of Savigny it 

 has been almost universally regarded as a sense-organ of some 

 kind — probably olfactory or gustatory, or in some way capable 

 of testing the quality of the inhalent current of water. The 

 reasons for this view have been : — 



1. The position of the organ at the entrance of the branchial 

 sac where a sense-organ would be of great apparent value. 



2. Its structure — a ciliated depression covered in part by 

 columnar cells, some of which closely resemble sense-cells. 



3. Its intimate relation with the ganglion, and the presence of 

 a nerve arising from the anterior end of the ganglion, running 

 towards the branchial aperture close past the dorsal side of the 

 tubercle, and presumably supplying it with nerves. 



In 1876 Ussow showed that the gland of unknown function 

 lying below the ganglion had a delicate duct, lined by cubical 

 epithelium, which ran forwards and opened into the tubular 

 posterior end of the funnel-like depression forming the dorsal 

 tubercle ; so that the slit of the tubercle was thus shown to be 

 merely the aperture of the duct from the neural gland. In 1SS1 

 lulin 1 confirmed this discovery, described minutely the condition 

 of the gland, the duct, and the tubercle in several species of 

 simple Ascidians, and declared that there was no connection 

 between the nerve running from the ganglion to the branchial 

 aperture and the tubercle, and that consequently the latter was 

 not a sense-organ, and was nothing more than the opening of 

 the duct. In a second paper, published shortly afterwards, Julin - 

 described the condition of these organs in two additional species, 

 and enunciated the theory, suggested to him by E. van Beneden, 

 that the neural gland was renal in function, and was the homo- 

 logue of the hypophysis cerebri of the vertebrate brain. In 

 favour of this homology may be con idered : — 



1. The position of the gland upon the ventral surface of the 

 nerve centre and above the pharynx. 



2. Its glandular nature. 



3. Its connection with the anterior end of the pharynx by a 

 duct — Balfour, Kolliker, and others having shown that the 

 hypophysis or pituitary gland in higher vertebrates arises as a 

 dorsal diverticulum from the stomodteum, but afterwards loses this 

 connection. 



From my own observations I can confirm Julin's statement as 

 to the presence of the duct from the neural gland and its con- 

 nection with the slit of the dorsal tubercle, and, like him, I am 

 una' ile to find any nerve supplying the supposed sense organ. 

 I have, however, in several cases seen certain of the epithelial 

 cells covering the edges of the slit which had a striking resem- 

 blance to sense-cells, such as those in the ectoderm of Actinia. 

 Thi= observation, taken along with Julin's descriptions, and espe- 

 cially with the condition of affairs in some specimens of Ascidia 



1 A rchivcs dc Biologic^ vol. ii. p. 59. 



2 Loc. cit. p. 211. 



mammillata which I have recently examined, has suggested to 

 me that possibly the dorsal tubercle may beeot/t the aperture of a 

 gland corresponding to the hypophysis and also a sense-organ, 

 probably of an olfactory or gustatory nature. 



Ascidia mammillata is one of the forms discu-sed by Julin in 

 his second paper. It is a large species with the branchial and 

 atrial apertures rather far apart, and the : a consider- 



able distance from the anterior end of the body. Julin found 

 that the neural gland in this species did not form the usual 

 compact mass, but was in a somewhat rudimentary condition, 

 and that besides having the usual duct running anteriorly to 

 communicate with the pharynx by the dorsal tubercle it had also 

 a number of short funnel-shaped apertures into the peri- 

 branchial or atrial cavity inclosed by the mantle ; so that in this 

 species the products of the gland might be excreted either into 

 the branchial sac (pharynx) or into the dorsal part of the peri- 

 branchial cavity, the region into which the intestine and the 

 genital ducts also open. 



In two specimens of Ascidia mammillata which I had an 

 opportunity of examining recently I found the neural gland in 

 exactly the condition described by Julin, but its duct had no 

 aperture into the pharynx, the dorsal tubercle being entirely 

 absent. The small tunnel-shaped apertures into the peribranchial 

 cavity were numerous and well developed, so that in the case of 

 these individuals the neural gland was connected with the 

 cloacal part of the peribranchial cavity only, exactly the arrange- 

 ment to be expected if the gland had a renal function. It seems 

 possible to me that this, or something like this, may have been 

 the condition of affairs in the primitive Chordata previous to the 

 point of divergence of the Urochorda. There may have been a 

 renal gland placed ventrally to the nervous system, not neces- 

 sarily at the anterior end only, and opening on the surface of 

 the body by one or more laterally-placed apertures, 1 this 

 gland being represented in the Tunicata by the neural gland, 

 and in the Vertebrata by the glandular portion of the pitui- 

 tary body. 



Then the dorsal tubercle apparently is or was a sense- 

 organ — possibly placed at first on the surface of the body, 

 since the anterior part of the pharynx develops from the epiblast 

 as a stomodaeum — and I think it probable that the connection of 

 the tubercle with the duet of the neural gland may be an after- 

 change, caused possibly by the enlargement of the pharynx into a 

 branchial sac, and the development of the peribranchial chamber. 

 It may readily be imagined how, as the result of the formation 

 of these cavities, the neural gland would be brought into closer 

 relation with the dorsal tubercle, and one or more of the funnel- 

 shaped ducts of the gland might, after having been carried in 

 from the surface by the formation of the lateral atrial involuii ins, 

 cone to open into the ciliated depression of the tubercle in place 

 of into the peribranchial cavity, thus producing very much the 

 condition described by Julin in his specimens of Ascidia mam- 

 tnillatt. By suppressing ihe original openings into the peri- 

 branchial cavity and leaving merely the secondary opening into 

 the pharynx by means of the dorsal tubercle, we arrive at the 

 c mdition found in all ordinary Ascidians. It is not easy to see 

 the reason for this change, as there is no apparent advantage to 

 be derived from it, but there is probably also no disadvantage, 

 since there is abundant communication between the branchial 

 sac and the peribranchial cavity through the stigmata or slits in 

 the wall of the former. 



This suggestion as to the origin of the present structure of 

 the neural gland and neighbouring organs in most Tunicata 

 implies that the pituitary body in the Vertebrata, which has lost 

 its connection with the ex'ei ir, and probably also its function, 

 has a similar history. In this view I am encouraged by some 

 remarks by Balfour,' from which it is clear that he considered 

 the pituitary body, judging from its development, to have been 

 originally a sense organ opening into the mouth, and possibly 

 corresponding to the Ascidian dorsal tubercle. He has also 

 suggested, 3 as an alternative, the possibility that the neural gland 

 in the Tunicata may be the homologue of the vertebrate 

 pituitary body. This is of course the theory supported by van 

 Beneden and Julin, and is open to the objection that it does not 

 account for the remarkable structure of the dorsal tubercle. The 

 view I hold combines both of those above mentioned by consi- 

 dering the pituitary body as the homologue of the neural gland, 



1 The lining of the peribranchial cavity, into which the ducts open in the 

 Ascidian. is derived from the epiblast, being formed in the embryo by a pair 

 of lateral involutions, which afterwards fuse dorsally. 



2 " Comparative Embryology," vol. ii. p. 359. 3 Loc. cit. p. 360. 



