635 
the Meiotic Divisions of Vicia Faba. 
but a light stain reveals that the chromatin retains its reticulate arrange- 
ment during the period of maximum contraction (Fig. 3). Later the cross 
connexions begin to break down (Fig. 4), and the formation of the free 
spireme is initiated. 
POST-SYNAPTIC PROPHASES. 
As the nucleus passes out of synapsis loops of free spireme are thrown into 
the nuclear vacuole (Figs. 5-9), the whole area of which, as the loosening pro- 
ceeds, becomes occupied by the chromatin thread ; this is not always wholly in 
the form of a simple spireme, but may still show the diamond-shaped mesh of 
the reticulum (P'ig. 10) to a greater or less extent. We have thus clear evidence 
of the persistence of a longitudinal split, which can be traced back through 
synapsis to the preceding telophase ; quite similar fissions may be recognized 
(Fig. 11) even where the cross attachments have already disappeared. 
Subsequently, after the chromatin has been spread through the 
nuclear vacuole, a second contraction begins, and, with the shortening and 
thickening of the thread, the longitudinal fission becomes difficult or impos- 
sible to trace. It is rendered clearly recognizable again only with the onset 
of the metaphase. 
The synaptic stages are frequently found, and usually extend throughout 
the whole anther ; they may be judged, therefore, to be somewhat prolonged. 
The same criteria indicate that the loosening and formation of the spireme 
take place slowly, but that the second contraction is a much more rapid 
process. The subsequent metaphases and anaphases, as well as the homotype 
division, appear to be swiftly accomplished. 
At all stages in meiosis, disintegration of the contents of the anther 
may set in. This is a not uncommon phenomenon among cultivated plants, 
and is indicated by contraction and abnormal appearance both of the 
cytoplasm and the nuclear contents. At an early stage, when the cell as 
a whole still presents a healthy appearance, globules of chromatin are 
extruded from the nucleus and pass into the cytoplasm, frequently invading 
a neighbouring cell (PI. XLIV, Fig. 33). They are for a time attached by 
fine threads to the parent nucleus, and closely resemble the ‘chromatic 
bodies’ described by Digby ( 1 ) as normally present in Galtonia candicans . 
The bodies observed in Vicia Faba , however, seem clearly related to the 
incidence of an abnormal condition. 
The Formation of the Gemini. 
The first indications of pairing appear among the free strands of spireme 
which have already been described as thrown out across the nuclear vacuole 
when the loosening of the synaptic mass begins. This occurs prior to 
the temporary obscuration of the longitudinal split, which is still quite 
clear some time after pairing has begun. Loops of similar form are observed 
lying side by side in pairs (Pl.XLIII, Figs. 5, 7, 8) in more or less close 
