Dixon — First Mitosis of ^jpore-mother-cells of Lilium. 3 



more or less parallel, and the chromomeres in them again become 

 visible. 



With these well ascertained stages before us it will be more easy 

 to discuss the connecting processes, entering the debateable gi'ound 

 from these firmly established facts. 



How the resting nucleus of the mother-cell of spores passes into 

 the dolichonema condition does not appear to have been very precisely 

 studied. It appears to be generally accepted that the irregular anas- 

 tomoses of the nuclear net-work disappear, and the chromatin granules 

 arrange themselves along the lengthening thread ^in series, while, 

 concomitantly, the nucleus greatly increases in size. ^ 



The stage described as "synapsis" intervenes between stages 1 

 and 2, e.g. between the dolichonema and strepsinema stages. Some 

 authors 2 believe that the balling together of the chromatin thread 

 observed in this stage is artefact, while Miss Sargant^ has observed it 

 in the living cells of Lilium. Synaptic nuclei are often found in very 

 carefully fixed material (both by Plemming's solution and alchohol) 

 and in close proximity to nuclei in the dolichonema stage. 



Apart from the question of the natural occurrence of synapsis a 

 difference of opinion exists as to how the strepsinema stage arises, 

 and is to be derived from the dolichonema stage. 



The great majority of observers describe a longitudinal division of 

 the chromatin thread during the end of the dolichonema stage. Accord- 

 ing to these writers the chromomeres in the thread divide first, and 

 then the linin carrying them .separates into two parallel filaments ; 

 these filaments then twist round each other, and diverging at places 

 give rise to the strepsinema condition. 



A careful study of the process, however, in the pollen-mother-cells 

 and embryo-sac of Lilium longiflorun led me^ to regard the strepsinema 

 condition as more probably arising from the looping on each other and 

 approximation of two portions of the dolichonematous thread. As 

 the actual transition from one stage to the other cannot be observed, 

 we can only indirectly infer the nature of the process from the appear- 

 ances of the successive stages presented in fixed specimens. The 



1 Guignard : " Le Developpement du Pollen," Arch. d'Anat. Mier., mars. 1899. 



2 Gmgnard, loc. cit., Schaifner: "The Division of the Macrospore Nucleus," 

 Bot. Gazette, 1897. 



3 Ann. of Bot., Sept., 1896. 



* Dixon : "On the Chromosomes oi Lilium lotigiflomm,'" Proc. Eoy. Irish Acad., 

 1896. 



