BULLETIN OF THE UNITED STATES FISH COMMISSION. 143 



toplasin iu which they are imbedded does, which appears to be a suffi- 

 cient proof that they are not the remains of the germinative vesicle, but 

 that they are merely vacuoles filled with fluid, such as are found in the 

 Protozoa, but which, unlike the spaces found in those animals, are not 

 rhjthmically contractile or pulsatile. 



ORIGIN AND FOEIHATION OF THE GERMINAL DISK. 



The germinal disk of the e^g of the mackerel measures one-fortieth to 

 one-fil'tieth of an inch in diameter, is biscuit-shai^ed, and is composed of 

 a light, amber- tin ted protoplasm several shades darker than the proto- 

 plasm which makes up the vitellus, which is remarkable for its glass- 

 like transparency. Normally, the disk always lies directly at one side 

 and apparently on the surface of the yelk, as indicated in Fig. 4 ; and 

 when the egg is in the water it is always immediately and exactly below 

 the latter. The disk may be developed independently of impregnation, 

 but in that case an embryo is never formed, and the egg soon disorgan- 

 izes, the vitellus collapses, and the whole protoplasmic mass, disk and 

 all, acquires a brownish, granular appearance, indicative of death and 

 disorganization. The disk takes its origin directly from the germinal 

 pellicle, which incloses the vitellus just like the rind which covers the 

 flesh of an orange. This layer at first thickens at one side, and its sub- 

 stance seems to flow gradually to the lower pole of the egg till the re- 

 sulting disk acquires the shape of a concavo-convex lens, with a thin, 

 sharp rim. Eventually the sharp rim disappears, it becomes smoothly 

 rounded at the edge, and the whole disk biscuit-shaped. 



It is probable that an extremely thin layer of protoplasm, which orig- 

 inally formed i^art of the germinal pellicle or layer, still covers the 

 ^itellus, and is continuous with the disk, and is synonymous with the 

 intermediary layer described by Van Bambeke and E. Van Beneden. 

 My reason for this statement is the fact that the disk is sometimes 

 found to have a thin layer of protoplasm extended outwards below its 

 thick rim over the vitellus, but which becomes so thin a little way out 

 from the edge of the disk that it becomes impossible to demonstrate it 

 without special methods. That the vitellus is covered by a structureless 

 membrane is proved by the fact that in the event of its being ruptured 

 its glairy contents will very rapidly escape, and its torn edges can be 

 seen limiting the opening or rent in its walls. The segmentation of the 

 thicker part of this membrane of i)rotoplasm next the disk is shown in 

 Fig. 4, and is also described at considerable length by Van Beneden in 

 the paper already referred to in a foot-note. In the crab-eater {JElacate) 

 a very peculiar wreath of flat cells is developed at the edge of the disk, 

 which, as in the ova of the su^jposed gadoid studied by Van Beneden, 

 appear to be continuous, with a layer of small cells below the thicker 

 lens-shaped part of the germinal disk. 



It may be well to retain the term intermediary layer for this structure, 

 but if the distinguished Belgian embryologist had succeeded in follow- 



