OF ARTS AND SCIENCES. 59 



so-called " parallel " cleavages (the third and the fourth) are as truly 

 meridian as the first and the second, using this term with reference 

 to the individual blastomeres, not with reference to the entire ovum. 

 There can be but little doubt that each blastomere, whatever be its 

 position, elongates during its division at right ayigles to its axis (as- 

 suming of course that each has poles of its own) ; and if this be 

 so, the difference between meridian and equatorial is one of name 

 only. In other words, a plane which may be called equatorial to the 

 entire ovum may be truly meridian to the individual blastomeres. 

 In regarding the " parallel " grooves as meridian, not to the ovum, 

 but to the blastomeres in which they occur, we have, it seems to us, 

 an explanation of these grooves that is in perfect accord with what 

 is now known in regard to cell-division, and escape the necessity of 

 appealing to an unknown factor, such as Rauber has had recourse to. 

 His "polflucht" theory, ingenious as it is, is built upon a hypothetical 

 tendency or force (polar repulsion ?), the existence of which is, in our 

 opinion, much more than doubtful. 



Irregularities of Cleavage. — The first cleavage splits the blastodisc 

 into two equal, or sub-equal blastomeres ; the second cleavage may 

 likewise be equal or sub-equal. The third cleavage, which usually 

 runs parallel to the first, dividing the blastodisc into eight blastomeres 

 arranged in two parallel and equal rows, sometimes results in an oval 

 instead of the usual rectangular form, in which one cell is central and 

 seven are marginal. During the entire season we found but one case 

 of this kind. The next cleavage gave five central and eleven marginal 

 cells. The cleavage was not followed further, but the ovum developed 

 into a perfectly normal embryo. 



Another very interesting variation in the 8-cell stage was met with 

 only twice. The third-cleavage planes were not parallel to the first 

 plane, but meridian (not only to the blastomeres but to the ovum), 

 resulting iu an oval figure with radial symmetry. The next cleavage 

 was concentric to the pole of the ovum. One of these was mounted 

 and the other was lost, so that we are unable to say how such a varia- 

 tion would end. The ova were supposed to belong to Ctenolabrus. 



On .July 7 an ovum belonging to Ctenolabrus was found a few 

 minutes before the formation of the first cleavage-amphiaster, and 

 watched as far as the IG-cell stage. The first three cleavages were 

 quite regular, but the position of the eight blastomeres was sutRciently 

 diff"erent from the normal, to induce a very unusual form of the 16-cell 

 stage. This stage presented seven central cells and nine marginal 

 ones. The ovum developed a perfect embryo. 



