KARYOKINESIS AND ITS RELATION TO FERTILIZATION. 1738 
plant-cells ; the former has recently stated that an absolute 
constancy does not exist, at least not in all kinds of cells. The 
generative cells exhibit in a remarkable manner the same 
number of loops (in one and the same species). (On this 
point cf. Boveri, p. 103.) 
Anyone who like Flemming believes only in a single thread 
in the foregoing stage, must regard it as dividing in the stage 
‘ of the “ loose skein ” into numerous pieces or loops. 
As already observed, Rabl also believes in such a division into 
several loops, but from the commencement he recognises numer- 
ous loops, so that for him the division is much more limited. 
After the “loose skein” there follows, as a third arrange- 
ment of the first stage of karyokinesis, the so-called “ seg- 
mented skein,” a term given to one of the most important of 
Flemming’s discoveries, and which is now known as a general 
phenomenon of karyokinesis, namely, the longitudinal splitting 
of all the thread-loops (see figs. 6 and 7). By means of the 
splitting of the several threads, as is proved by later events, a 
resolution of the whole chromatic mass of the nucleus into 
two equal parts takes place, and the succeeding phenomena 
result in the portions separating from one another, and 
grouping themselves as two new daughter nuclei. 
Rabl says definitely that he has always found the longi- 
tudinal splitting of the chromatic threads to be completed at 
the end of the skein phase. E. van Beneden (for Ascaris 
megalocephala) lays special stress on the fact that the two 
daughter-threads are exactly alike down to the smallest detail. 
The longitudinal splitting, according to him, is at first not 
complete, so that the sister threads for some time remain 
connected with each other at their two ends by means of a 
less deeply staining substance. This connection still continues 
when the sister-threads separate towards the two poles. There 
are then seen, as was first shown by van Beneden and later by 
Rabl, fine achromatic threads (‘ filaments réunissants ” of van 
Beneden) stretching from the ends of the separating loops, 
which for some time connect the chromatic loops of the 
daughter-nuclei (‘‘ Dyaster*’). These threads must be clearly 
