Pollen Development in Lactuca. 397 
Fig. 2. Loculus at later stage. Pollen mother-cells in postsynaptic spireme stage. One 
mother-cell is binucleate. Tapetal cells with from one to four nuclei, x 1,500. 
Fig. 3. Loculus at similar stage to Fig. 2, one tapetal cell showing two nuclei in synizesis. 
x T ’5°°* 
Fig. 4. Portion of loculus showing variation in shape and size of tapetal cells at opposite sides 
of loculus, x 1,500. 
Fig. 5. Pollen mother-cell with nucleus in resting condition, x 3,000. 
Fig. 6. Pollen mother-cell showing the beginning of contraction of the reticulum at one point, 
x 3,000. 
Fig. 7. Slightly later stage than Fig. 6. The nuclear membrane has broken down at one side. 
First indication of transformation from reticulum to spireme, x 3,000. 
Fig. 8. Pollen mother-cell nucleus showing final stage of contraction. Membrane plainly 
visible, surrounding the reticulum, x 3,000. 
Fig. 9. Similar stage to Fig. 8, but membrane less apparent. Reticulum still connected to 
nuclear membrane by numerous fine threads, x 3,000. 
Fig. 10. Similar stage showing both reticular membrane and connecting threads, x 3,000. 
Fig. 11. Pollen mother-cell nucleus in typical synizesis stage, x 3,000. 
Fig. 12. Pollen mother-cell nucleus showing spireme emerging from the synaptic knot, x 3,000. 
Fig. 13. Similar stage to Fig. 12, but showing cytomyxis. Extrusion of chromatic material 
has taken place from the smaller cell on the left, x 3,000. 
Fig. 14. Binucleate pollen mother-cell, both nuclei in presynaptic resting stage, x 3,000. 
Fig. 15. Binucleate pollen mother-cell. Both nuclei about to form synaptic knot, x 3,000. 
Fig. 16. Binucleate pollen mother-cell, nuclei in spireme stage, x 3,000. 
Fig. 17. Binucleate pollen mother-cell, probably postsynaptic. x 3,000. 
PLATE XVII. 
Fig. 18. Early postsynaptic spireme stage, thread nearly continuous, a few free ends visible, 
nuclear membrane indistinct, x 3,000. 
Figs. 19-26. Various stages showing looping of the portions of the spireme to form finally 
nine bivalent chromosomes, x 3,000. 
Fig. 27. Slightly later stage; portions of the spireme, in the form of loops, having broken off. 
x 3,000. 
Figs. 28, 29. Further stages of the segmentation of the spireme. Nine segments can be counted, 
x 3,000. 
Fig. 30. Segmentation completed. Note the four long twisted bivalents and five shorter and 
more condensed, x 3,000. 
Figs. 31-36. Further stages of the progressive condensation of the nine bivalent chromosomes, 
showing also the torsion, x 3,000. 
Fig. 37. Diakinesis with ten chromosomes. The two marked ‘ a ' probably represent the two 
halves of a single original bivalent chromosome, x 3,000. 
F'igs. 38, 39. Diakinesis. Nearly all trace of the bivalent nature of the chromosomes is lost, 
x 3,000. 
Fig. 40. Diakinesis. One bivalent chromosome almost split into its component halves, 
x 3,000. 
Figs. 41-44. Diakinesis. Nine bivalent chromosomes, x 3,000. 
Fig. 45. Diakinesis. Note threads connecting two pairs of bivalents, x 3,000. 
Fig. 46. Diakinesis. Two of the longest bivalents fused end to end. x 3,000. 
Figs. 47, 48. Diakinesis. Only eight bivalents in each, no definite evidence of fusion, x 3,000. 
Fig. 49. Single bivalent chromosome, showing the loops, x 3,000. 
Fig. 50. Heterotypic metaphase in polar view. Only seven independent chromosomes, but 
those marked 1 a' represent fused pairs of bivalents, x 3,000. 
Fig. 51. Heterotypic metaphase in polar view. Eight chromosomes, one clearly representing 
two bivalents fused end to end. x 3,000. 
Fig. 52. Metaphase showing nine bivalents. Note points of attachment to spindle fibres, 
x 3,000. 
Figs. 53, 54. Metaphase with eight bivalents. Note again points of attachment, and the ring 
of dark staining material in Fig. 54. x 3,000. 
D d 
