FORMATION OF THE GERMINAL LAYERS IN TELEOSTEI. 223 



rings urn den Rand des Keimliiigels Kerne auftreten, die in ganz 



regelmassiger Weise angeordnet sind. Es sind wasserklare, runde Blaschen, 

 ohne irgend welche Komchen im Innern, die in concentrischen Kreislinien, auf 

 das Centrum des Keimhiigels bezogen, sich gruppiren." " Man sieht namlich 

 zwischen den blaschenartigen Kernen zarte Contouren auftreten, die genau an 

 einander schliessende polygonale Felder umgrenzen, deren Mittelpuncte die 

 Kerne einnehmen. Kurz es entsteht eine Lage eines regelmassigen, aus 



hexagonalen Zellen gebildeten Platten-epitheliums." " Unter den sich 



furchenden Keime ein besonderes Blatt sich bilde, das nicht aus den Fur- 

 chungszellen herzuleiten ist, denn die Zellen desselben entstehen frei in einer 

 den Dotter bekleidenden diinnen Blastemschicht, indem als Erstes die Kerne 

 derselben erscheinen." " Ob dieses Blatt wirklich zum Darmdriisenblatt wird, 

 muss dahingestellt bleiben, vielleicht ist es nur eine voriibergehende Bildung, 

 was aber wohl unwahrscheinlich." 



So far as I am aware, this is the first clear account of a nucleated zone of 

 protoplasm outside the segmented disc in Teleostean fish ovum. It is true that 

 at an earlier date Lereboullet (21) observed a similar layer of cells in the 

 egg of the pike, which he concluded was transformed into the lowest 

 germinal layer. He was, however, of opinion that the cells were derived 

 from small yolk spheres (globules vitellins), and failed to recognise the inde- 

 pendent formation of the nuclei, which by Kupffer were ascribed to " free cell 

 formation." 



So far as the Teleostei are concerned, the existence of this layer has since 

 been thoroughly established. The points on which more recent investigators 

 differ are firstly — the source from which the nuclei in this layer are derived; 

 and secondly, the ultimate fate of the cells derived from it. 



Ten years ago Klein (16) made an important contribution to the subject, 

 and reviewed the position taken up by earlier authors. This author concluded, 

 from a study of the early stages of the trout, that besides the blastoderm 

 proper, it is necessary to study closely the behaviour of the subgerminal and 

 paragerminal substance, which " bears at all stages of development an im- 

 portant genetic relation to the blastoderm and the embryo." Klein calls the 

 segmented portion of the blastoderm archiblast, and the unsegmented portion 

 in connection with the yolk is the parablast. I have followed Klein's nomen- 

 clature in the present paper, although, as already pointed out, the terms are not 

 used in the sense originally applied to them by His. The parablast layer was 

 first correctly noted by Oellacher (23), who described it as continuous with 

 the germinal disc in early stages, but mistook it for a vitelline membrane. 

 Klein restricts the term parablast to the thickened welt of protoplasm, having 

 a somewhat triangular section which forms a rim round the segmented blasto- 

 derm in early stages. Van Bambeke (3) has described a similar layer in 



VOL. XXXIII. PART I. 2 H 



