354 W. J. C. Lawrence Cytoiogia 2, 



this species. Owing to the smallness of the chromosomes in Dahlia, 

 prophase is not as a rule suitable for a critical study, therefore the ob- 

 servations presented here have been made mainly on the stages from 

 early diakinesis to second telophase. 



Prophase 



At zygotene the single threads are seen lying side by side in 

 various degrees of approximation, and at diplotene loops are formed 

 with several points of contact. Examination of early diakinesis shows 

 that the 36 chromosomes invariably form 18 pairs which are approxi- 

 mately equidistant from one another around the periphery of the 

 nucleus. 



Di akinesis 



In over 50 nuclei examined at diakinesis, only 4 cases were found 

 of possible associations of more than two chromosomes. In 3 of these 

 cases it was not possible to determine whether the associations were 

 real (i.e. whether connections existed between the component members) 

 owing to the unfavourable orientation of the chromosomes in the plane 

 of vision. The fourth case was an apparent chain quadrivalent, but it 

 was adjacent to the nucleolus, in the vicinity of which fixation is always 

 poorest. No ring quadrivalents have ever been seen, and if the 4 cases 

 mentioned above were quadrivalents, then from the disposition of the 

 members they must have been chain quadrivalents. In that case their 

 frequency is less than 1 to 200 bivalent associations. 



The bivalents are of two kinds : in the first kind the members are 

 connected at one end only : in the second there are connections at 

 both ends. At the beginning of diakinesis the pairs lie side by side 

 and are somewhat elongated, but as diakinesis proceeds the chromatin 

 contracts and a marked repulsion of all the chromosomes is seen. 

 This results in the swinging apart and subsequent orientation of the 

 component members of the type with a single terminal connection so 

 that the two chromosomes are apparently paired " end-to-end." These 

 observations are probably characteristic for all small chromosomes 

 with terminal chiasmata. 



A consideration of the work of Darlington on plants with large 

 chromosomes shows that his conclusions in relation to the function and 

 form of chiasmata provide a satisfactory explanation of the behaviour 

 of the small chromosome of Dahlia. Chiasmata observed at diplotene 



