146 Gates, Tetraplokl Mutants and Chromosome Mechanisms. 



of "good" grains, and of these 62 per cent, were 3-lobed, 87 90 

 per cent. 4-lobed, while about 7 per cent, had 5 or more lobes. 

 But one individual of this race was distinctly smaller in flowers, 

 leaves, etc. possessed about 33 per cent, "good" grains, and of 

 these 23 per cent, were 3-lobed, 75 per cent. 4-lobed, and about 

 2 per cent. 5- or more lobed. This individual probably contains 

 less than 28 chromosomes. In the narrow-leaved gigas above men- 

 tioned, the pollen was all sterile, but 3-lobed and 4-lobed grains 

 were both numerous. 



In an individual of gigas Sweden examined, the amount of 

 "good" pollen was found to be about 37 per cent., of which 9 per 

 cent, were 3-lobed, about 88 per cent. 4-lobed, and about 3 per 

 cent, with more than four lobes. 



Since the triploid mutants also possess a certain (unknown) 

 percentage of 4-lobed grains, it is obvious that there is a general 

 parallel between the number of chromosomes and the percentage 

 of quadrangular grains. 



The percentage of quadrangular and triangular pollen grains 

 is thus a very useful preliminary criterion to the approximate 

 number of chromosomes possessed by a plant. It seems safe to 

 conclude that all quadrangular grains contain more than 7 chromo- 

 somes, though it remains to be proven that they all possess the 

 full 14. Probably triangular grains with an extra lobe possess an 

 intermediate chromosome number. 



It is shown by calculation from previous measurements, that 

 in the pollen mother cells the ratio of increase of the cytoplasm in 

 gigas as compared with Lamarckiana is only 1.5 : 1 while in the nuclei 

 the ratio is 2.16 : 1. The deficiency of cytoplasm in the gigas pollen 

 mother cells therefore amounts to about one-quarter and it is sug- 

 gested that the extra lobe of the gigas pollen grain serves to restore 

 the normal karyoplasmic ratio. 



6. Several characters of 0. gigas which were cited by de Vries 

 as changes occurring independently of the chromosome-doubling, 

 have been shown to be readily explainable as a direct result of 

 the tetraploid condition with its larger cells and nuclei. These 

 characters include (1) the strong biennial habit (2) the larger seeds (3) 

 the short fruits (4) the greater susceptibility to frost. Whether any 

 characters (such as leaf-shape) remain which can not be explained 

 in this way, is a difficult question to decide in the present state 

 of our knowledge of morphogenetic processes, but the effort should 

 evidently be made, to explain the characters of giant types as far 

 as possible on the basis of a single primary change, whether this 

 be in chromosome number or merely in cell size. 



7. A comparison of the observations of Geerts on meiosis 

 in the pollen mother cells of O. lata X gig as i with my own, shows 



