12 DR. J. F. GEMMILL ON THE DEVELOPMENT OE 



subequality may also be present in healthy ova. One also comes across cases in which 

 the early segmentation is very markedly unequal. Such ova may seem to develop 

 normally during the gastrula and the early larval stages, but none of them which I 

 isolated entered on the later larval stages or underwent metamorphosis. 



Mastula and Gastrula. — The process of blastula and gastrula formation presents 

 those remarkable features which were first carefully described by Masterman (i8, 

 pp. 378-381) as occurring in the ova of Cribrella. 



A rudimentary primary blastoccele in the form of narrow slit-like spaces develops 

 within the cell-mass. Numerous small furrows running into one another next appear 

 all over the external surface, producing a finely lobulated appearance almost as if the 

 egg were reverting to its middle segmentation-stages. Soon the furrows become fewer 

 but better marked. This takes place through the levelling up of some and the 

 confluence and deepening of others. The surface pattern now becomes simpler, 

 simulating an earlier stage in segmentation. The resemblance is often very striking, 

 even as far downwards as the 16-, the 8-, and the 4-celled condition. Usually, how- 

 ever, instead of the knobby segmentation pattern, there is an appearance as of lobes 

 running together, which may best be described as resembling the convoluted surface of 

 a cerebral hemisphere. In the end, all the furrows disappear, and the surface becomes 

 perfectly smooth, at least as seen from above. Sections of eggs at the time when the 

 fiuTows first appear show the two following peculiarities: (1) the superficial cells are 

 elongating and arranging themselves in a single layer at the surfaces between the 

 furrows ; (2) at the same time the deeper cells of the blastula are undergoing a similar 

 rearrangement with reference to the sides and bottoms of the furrows. It is obvious 

 that the result will be to produce, out of the originally almost solid eel] -mass, a single 

 greatly folded layer of columnar cells (PI. I. fig. 5) enclosing an irregular and slit-like 

 central space which may be called the definite blastoccele. A few cells, not falling 

 into line with the others, remain within this cavity and form the earliest mesenchyme. 

 The disappearance of the first furrows and the consequent simplification of the 

 surface pattern is accompanied by further narrowing and elongation of the individual 

 cells. In this way the blastula-wall, although it is losing in superficial extent, is 

 still able to accommodate all the cells which first entered into its formation. There 

 is now also a slight increase in the size of the blastula as compared with the 

 unsegmented egg, the difi"erence amounting to something like one-eighth of the 

 diameter of the latter. 



It is worthy of note that during all these changes the paler colour of the lower pole 

 is preserved. Cilia make their appearance on the surface about the same time as the 

 first blastula-furrows, causing currents within the membrane of fertilisation which still 

 persists. This membrane now disappears and the blastula begins to perform irregular 

 rotary movements in the water, its paler side always remaining underneath. 



Gastrulation. — Gastrulation takes place by invagination and commences before the 

 last surface-furrows have straightened out. Indeed, it may be described as beginning by 



