EAELY STAGES OE SEGMENTATION OF ECHINUS HYBEIDS. 503 
pi’ojection extending’ towards the line of cell-division; in 
others the daug’hter-nuclei are normal, bat a stained mass of 
chromatin remains in the cytoplasm on the line dividing the 
two cells. In 2-cell stages such a chromatic mass, usually 
very small, but of varying size, is very commouly found 
pressed against the faces by which the cells are in contact, 
sometimes in only one of the daughter-cells, at others in both. 
In other eggs no sign of elimination is found, but it is possible 
that in some at least of these the eliminated chromosome has 
been absorbed. 
The second division figures in our material are mostly in 
metaphase or early anaphase, so that it is difficult to deter- 
mine with certainty whether any chromosomes are eliminated 
in them; in one casein an anaphase one chromosome appears 
not to be dividing (miliaris $ x acutus but in the 
few other cases observed no evideiice of elimination in the 
second segmentation division was found. The spindles in 
these crosses, especially in the second division, are very short, 
with poles near together; not very infrequently nuclear divi- 
sion occurs not followed by cell-division. 
In general, then, it ma}' be concluded with regard to the 
miliaris crosses, that in (1) acutus ? x miliaris a 
small but variable number of vesicles, such as we have 
described in acutus ? x esculent us d, is sometimes 
formed, and that the chromosome number is possibly reduced 
thereby to a small extent. (2) In esculentus $ x 
miliaris there is no abnormality, and the observed 
chromosome number is that which is to be expected from our 
determination of the numbers in the parent species. (3) In 
the converse crosses miliaris ? x acutus and escu- 
leutus c?j “o vesicles are produced, but in the anaphase of 
the first division one or moie chromosomes commonly either 
fail to divide, or divide so late that they are not included in 
the daughter-nuclei. 
