i9i<5] NOTHNAGEL— REDUCTION DIVISIONS 463 



meshes. While the resting nucleus is said to consist of diamond- 

 shaped meshes, it is in all probability the same as the ladder-like 

 formation of Allium tricoccum. 



As has been .previously stated, the number of chromosomes in 

 Allium tricoccum is small and the size large, so that the difficulty 

 so often encountered in following the development is considerably 

 lessened. Throughout these critical phases every precaution has 

 been taken to prevent overlooking important stages. Since the 

 pollen mother cells at the upper end of a loculus are a little earlier 

 than those at the lower end in development, at least two consecu- 

 tive stages could be found in a single section. In every case the 

 later of the two stages observed has been found in the upper end 

 of the loculus, where the third has been found and drawn, decreasing 

 to a great extent the possibility of omitting critical stages or placing 

 the wrong interpretation upon the origin of the double character 

 found in the resting nucleus. Had figs. 5, 7, or 8 been the first 

 nuclei observed after the telophase of the previous division, the 

 conclusion could readily be drawn that the double thread arose 

 by the pairing of somatic chromosomes; but, after seeing the 

 beginning of vacuolization (figs. 1, 2) and following its develop- 

 ment step by step (figs. 2-9), no other conclusion is possible 

 than that the paired threads going into synapsis are the two 

 halves of single somatic chromosomes, and not paired somatic 

 chromosomes, as held by Gregoire (16) and Yamanouchi 

 (32, 33). 



Fig. 8 illustrates the breaking down of the connecting strands 

 between the two sides of the ladder-like structures, and on the 

 right side of this figure the act has been completed, leaving the 

 halves completely separated except for the portion connecting them 

 at the end. Mottier (22) states that " the delicate threads joining 

 the chromatic masses may be found lying close to each other and 

 parallel, but this does not signify that a double spirem is in process 

 of formation"; although in his figures of this stage (Mottier 22, 

 figs. 1, 17, 34) he shows nuclei very similar to that in fig. 5 of 

 Allium tricoccum; and by tracing this further it is very probable 

 that the origin of the double nature of the hollow spirem, as 

 described by Mottier (21, 22, 23), might be found. 



