The spermatogenesis of Peripatus (Peripatopsis) balfourî. 299 



3. Synapsis Stage. 



The term "synapsis" is used here as it was first employed by 

 Moore (1895), and later used by myself (1898), to denote that portion 

 of the spermatogonic anaphase in which the reduction of number of 

 the chromosomes is etfected. There is no sharp line of demarcation 

 between this stage and what has been termed by me the "early 

 anaphase"; for the sake of convenience in description, the synapsis 

 stage may be said to begin when the nuclear membrane has appeared, 

 and when in the central pole of the nucleus, thus bounded, a large 

 amount of nuclear sap is present. 



The chromosomes are now elongated, with irregular surfaces 

 (Fig. 63, Plate 19; Figs. 64—82, Plate 20); their distal ends are 

 held close to the distal portion of the nuclear membrane by their 

 attachment to the connective fibres. Of mantle fibres there is no 

 longer any clear trace. 



The phenomena of importance in the synapsis stage 'are the 

 joining of the original 28 univalent chromosomes into pairs whereby 

 14 bivalent chromosomes are found, and the longitudinal splitting of 

 all the chromosomes. In most, if not all the cases of spermatogenesis 

 hitherto described the chromatin loops form a dense, practically un- 

 decipherable coil in the synapsis stage ; but in Peripatus this grouping 

 of the chromosomes is much less dense, and their form and arrange- 

 ment can be determined with relative ease. 



As we have shown above, the beginnings of the formation of 

 chromosome pairs can sometimes be observed as early as the 

 preceding stage (Plate 19, Figs. 49—52), but the details of the 

 process are most clearly marked in the synapsis. The chromo- 

 somes lie either more or less parallel to, or else their axes make 

 an angle with, the axis connecting the central and distal ends of 

 the cell. The frequently observed curving of the distal ends of the 

 chromosomes can well be ascribed to their attachment to connective 

 fibres which have been pushed axially by the constriction of the cell 

 body (Figs. 60 — 63, Plate 19), which movement could well cause a 

 corresponding movement on the part of the distal ends of the chromo- 

 somes. 



But it is at the central ends of the chromosomes that the 

 movements of significance for an understanding of the phenomena of 

 the reduction of the number of the chromosomes are taking place. 

 The central ends of every two of the univalent ^8 chromosomes be- 



