438 
BULLETIN OF THE UNITED STATES FISH COMMISSION. 
Kupffer, Van Beneden, Henneguy et al. believe that cells are added to the germ 
from the periblast without stating that the cells will play any definite role. Klein 
and Van Bambeke believe that the entoderm is derived from the periblast. Hoffmann 
(1883) formerly believed that the periblast had nothing to do with the formation of 
the blastoderm; but later (1888) he inclined to the view that the entoderm with all 
its derivatives comes from the periblast. 
Kowalewski (1886) states that the free nuclei increase in size, and finally disin- 
tegrate. The function (p. 455) of the periblast he considers “ eine ernahrende und seine 
Bedeutung die eines provisorischen Organes das nach der Beendigung seiner Function 
zu Grunde gehet.” Hoffmann (1883) and Ziegler are of similar opinion. In the goldfish 
Kowalewski found tbe periblast to arise below the whole of the blastoderm. In Poly- 
acanthus viridiamatus they arise only at the margin. 
Wenckebach (1886) in studying Belone found that the periblast nuclei arise as early 
as tbe 64-cell stage, and that they increase at the margin of the blastoderm both by 
fission and by the disappearance of the cell walls of two or three series of cells. The 
same conditions he finds in Perea fluviatilis. 
In another egg (the largest pelagic egg, 4 mm., a Stolepliorus) a number of cells 
of the lower side of the blastoderm swell, become loose, assume an irregular form, and 
fall to the bottom of the segmentation cavity and unite with the periblast. The 
periblast in this case arises at the base as well as the margin of the blastoderm. A 
similar condition was also found by Oellacher, (Z. W. Z. xxiii, 12) in the brook trout. 
Wenckebach further found that the periblast nuclei undergo a slow degeneration (p. 
231) and have no part in the formation of the embryo. They become enlarged and 
in colored preparations show an irregular, large-meshed structure which is evidently 
due to the formation of vacuoles in the nucleus. They no longer divide. The ultimate 
fate of the periblast nuclei is degeneration and absorption. He found the residue 
after yolk absorption rich in protoplasm in which the nuclei were heaped. They 
become irregular in outline, structureless, and finally merge into one mass before 
final absorption. He is not certain as to their function. 
Hoffmann (1888) states that the third segmentation in the salmon is horizontal and 
separates the lower four periblast cells from the upper. At this stage the lower cells 
have distinct lateral boundaries. As soon, however, as the first merocytes (cells 
derived from the periblast) are formed the margins of the periblast cells are lost. 
Through equatorial cleavage new merocytes are continually produced by the free 
nuclei, while through meridional cleavage the number of free nuclei which remain in 
the plasma of the yolk is increased and they soon extend over the yolk beyond the 
margin of the blastoderm. As soon as a large number of merocytes appear the peri- 
blast nuclei are transformed into bodies which may be larger than the overlying cells. 
McIntosh and Prince (1890) also give a long discussion of the various theories of 
the origin and meaning of the periblast, without, however, touching the vital points. 
Wilson (1891) finds that the periblast nuclei become greatly vacuolated and that 
their outlines become irregular. The physiological use of the periblast he does not 
know. He supposes that the periblast is finally absorbed by the liver. 
The periblast seems to me to have been studied and discussed a great deal more 
than its importance warrants and I shall confine myself to describing the actual con- 
ditions found in Cymatogaster. I may only say that Agassiz and Whitman’s conclusions 
as to the origin of this layer are borne out by this fish, and the accounts, when 
