Studies on Synapsis. 



61 



premeio tic divisions. . ." After this unequivocal statement the interpretation 

 given to account for the reduction of the chromosomes is that " the spireme 

 threadwork tends to separate out into half as many lengths " as the number 

 of premeiotic chromosomes. Now, it may safely be said that in general (as 

 well as in the case of Periplaneta) recent research has fully justified Bolles 

 Lee in his contention that "ni la telophase, ni aux prophases, ni k aucun 

 moment de I'existence de ces noyaux, il y ait formation d'un spireme 

 continu." Yet even recent workers, like Arnold, 1909 (1), Nakahara, 1919 

 and Harman, 1920 (2), in their inability to account for telosynaptic union by 

 other means, fall back upon the conception of a spireme segmenting into the 

 haploid number of threads as a means of effecting reduction. 



The importance of an understanding of the various phases characteristic 

 of nuclear division was emphasised in the last study in reference to com- 

 parisons made between the meiotic processes in Osmimda and Periplaneta. 

 Attention was there directed chiefly to the question of whether the telo- 

 phasic chromosomes divided in anticipation of the succeeding mitosis. A 

 preUminary examination of spermatogenesis in Libellula in May, 1920, 

 indicated that the material was of a very favourable kind for a more 

 searching study of nuclear organisation in the premeiotic phase ; and the 

 present communication contains the results of observations made on material 

 collected at that time and subsequentl}\ The nymphs were identified, by the 

 courtesy of Mr. H. Campion, of the British Museum, as L. depressa. 

 Partic\alars of the materials and methods employed will be given with the 

 observations recorded. 



Previous Wai'k on Odonata. 



Since the Odonata, as a group, have attracted very little attention hitherto 

 from cytologists, a brief summary of previous work may first be given. The 

 only observations recorded with respect to the chromosome cycle are those of 

 McGill, Lefevre, and Smith. 



McGill, 1907 (3) has published an accoimt of the oogenesis of two genera, 

 Anax and Plathemis ; the more salient particulars are : (a) the presence of 

 yolk nuclei of the Lumbricus type in the young oocyte ; (6) the condensation 

 of the " synaptic spireme " around the plasmosome to form a double nucleolus ; 

 (c) the emission of fluid particles from the double nucleolus during the 

 growth period of the egg. The author speculates that these latter are 

 " reprecipitated " after " undergoing chemical change " to form the chromatin 

 of the mature oocyte. The alleged condensation of the " synaptic spireme " 

 around the plasmosome is homologised with synizesis in the male sex cells ; 

 and the views of Eetzius, Leydig, Strassburger, Guiguard, and Mortens, who 

 regarded the nucleolus as a temporary form of chromatin storage, are believed 



F 2 



