July 14, 1887] 



NATURE 



247 



I 



which I owe to the generosity of my former teacher, 

 Prof. Howes. 



Before giving the detailed account of my investigations, 

 I may say that neither the anatomical nor developmental 

 studies so far made, give any direct clue to the origin of 

 the organ. 



That which seems to me the most likely hypothesis I 

 shall give at the end of this paper, and in its favour I can 

 at least say that it is a morphological explanation of the 

 evolution of the parietal eye, which, so far as I know, is 

 not inconsistent with any known facts. 



The epiphysis in Ammocoetes has been described by 

 Ahlborn {Zeitsch.f. 70tss. Zool., Bd. xxxix.). His descrip- 

 tion is mainly correct, and but little can be added to it. 

 The epiphysis itself is divided into a dorsal and a ventral 

 vesicle, and as we are not concerned here with the ventral 

 one, I shall ignore its existence. 



In large Ammocoetes the dorsal vesicle lies deep under 

 the skin, and far removed from the light ; its position 

 being marked externally by a clear white spot just behind 

 the opening of the nose. 



It is a simple closed sac, and retains its attachment to 

 the brain. The dorsal wall is thinner than the ventral, 

 and is made up of a layer of flattened cells, which are 

 not modified to form a lens. 



The ventral wall is a much more complicated structure. 

 Towards the inside of the vesicle it presents a layer of 

 rod-like cells, which are more like the rods of a retina 

 than like anything else. Externally (with regard to the 

 vesicle) to this layer are two or three irregular rows of 

 nuclei. There is no lens and no pigment, except a few 

 very minute dots. 



In this stage the retina of the parietal eye of Ammo- 

 coetes somewhat resembles that of Cyclodus, figured by 

 Spencer, but is somewhat better developed, and tends 

 towards the condition found in Vai-anus gii^nntcus. 



Except in the presence of the minute dots of pigment, 

 and in the fact that the dorsal wall of the vesicle is not 

 connected by fine strands with the ventral wall, as 

 Ahlborn supposed, there is nothing new in this descrip- 

 tion, and even now we cannot say that the parietal eye 

 of fully-grown Ammocoetes is very highly developed. 



In the adult Petromyzon, just as the paired eyes are 

 highly developed so also do we meet with an increased 

 development of the parietal eye. As is well known 

 since Wiedersheim's researches, the brain of the adult is 

 much compressed in an antero-posterior direction. The 

 dorsal vesicle of the pineal gland lies much further for- 

 wards, and more dorsally than in the larva, so that it 

 comes to be nearer the external surface of the body, while 

 it lies buried in the roof of the skull. Its posterior wall is 

 densely pigmented, so much so that it is impossible by 

 ordinary means to make anything out of the structure of 

 the cells composing it. These points can be seen very 

 plainly in longitudinal vertical sections through the brain 

 and skull (see figure). 



I ought to mention that the clear white patch of skin 

 lying above the organ is much larger and more marked 

 than in the Ammocoetes. It is, however, difficult to 

 suppose that the white patch is here of much physio- 

 logical importance, and it can only be referred back to a 

 time when the eye in Petromyzon was of more use than 

 at present. The anterior wall is composed of cells which 

 are thrown into folds (possibly in part due to contraction) 

 projecting into the cavity of the vesicle. 



I mentioned above that in the full-grown Ammocoetes 

 there are only a few minute dots of pigment present. So 

 few and so small are these, that unless specially sought 

 for they would be overlooked, as indeed they have been 

 by previous observers. The state of things is much 

 different in the young Ammocoetes of about 2 inches 

 in length. There, as in the adult, the retina of the 

 parietal eye contains a large deposit of pigment. This 

 was first shown me by Dr. Schwarz (a pupil of Prof. 



Weismann's), who has made, for the study of the pairei 

 eyes, some very fine sections of very young Ammocoetei; 

 at stages which I had failed to obtain. I shall figur 

 these sections in the complete account I have in prepara 

 tion. In the young Ammocoetes the parietal eye is large 

 and exceeds in size either of the paired eyes. Itsposterio 

 wall is really a well-developed retina, with long rod lik 

 elements embedded in pigment, and a series of oute 

 layers of spherical nucleated bodies. Its anterior wal 

 consists of several layers of rounded cells, but it doe 

 not form a lens. 



In the specimen of Petromyzon marinus mentionei 

 before, owing to the soft state of the brain I could onl 

 make out a very deep fossa in the skull in the position ii 

 which the " eye " is situated in P. planeri. The whit( 

 patch of skin is here very large indeed, and on the who! 

 I am inclined to think that the parietal e) e in Pctrojuyzo, 

 marinus would well repay further investigation. 



In Myxine the state of things is even more surprising 

 Here the parietal eye is a large flattened vesicle lying 01 

 the brain and connected with it by a very short soli( 

 stalk. There is externally no white patch of skin, bu 

 lying in the skin above the vesicle there is a flattenec 

 body, which, in structure and position, more nearl 



Longitudinal vertical (sagittal) section through the parietal eye of an adul 

 Petromyzon planeri [Zeiss C. oc. 2 cam.], br, brain; c.t., connectiv 

 tissue; n, position of nose; p.e., pigment of the retina; P.O., parieta 

 eye, i.e. dorsal vesicle of the epiphysis; r, retina; s.d.p., subderma 

 pigment ; sk, skull ; sn, skin ; v. v., ventral vesicle of pineal gland. 



resembles the " Stirn-driise '' of Amphibians than any 

 thing else. This " Stirn-driise," as is well known, is's 

 rudimentary portion of the epiphysis, and hence of the 

 parietal eye. 



There is no lens and no pigment in Myxine. The 

 anterior wall of the vesicle consists of a single layer o; 

 somewhat flattened cells. 



The retina has essentially the structure of that of the 

 parietal eye of Varanus, but it lacks the pigment which 

 is there present {vide Spencer, "Pineal Eye in Lacertilia,' 

 Q.J.M.S., vol. xxvii. Part 2, Plate XIV. Figs, i and 6). 



Bdellostoma seems, in this, and, as was first shown bj 

 Johannes Miiller, in other points in the structure of its 

 brain, to resemble Myxine. Without discussing the 

 matter at length, I may say that in the parietal eyes ol 

 Petromyzon and Myxine we have to deal with structures 

 which are still well developed, and which were probably 

 once much more developed than now. In this connexion 

 the history of the changes in Ammocoetes is very inter- 

 esting, and all the more so as confirming and ex- 

 tending Dohrn's opinion that the Cyclostomata have 

 degenerated from highly developed fishes. The parietal 

 eye in Ammocoetes, like many other of its organs, makes 

 a good start, and only degenerates as the Ammocoete 

 degenerates. When the Petromyzon state is reverted to, 



