FRAGMENTS AND RINGS 73 



problem of its adaptation to meiosis will be considered later (Ch. V). 

 In some species an enormous range of size is found in the chromo- 

 somes of the normal complement (v. Section 4 (ii) ), and great 

 constancy is sometimes also found in the number of fragments in an 

 old clone. Thus in a Tradescantia plant with 5 fragments, 4 and 6 

 were found exceptionally while in a Fritillaria imperialis plant with 

 6 fragments, 12 were found in an anther, evidently owing to both 

 halves of each of the 6 going to the same pole in a mitosis, but 

 otherwise the number was constant. In other species, behaviour is 

 much less regular. In a group of Ranunculus acris plants, probably 



Fig. 18. — Metaphase of mitosis in root-tip of abnormal plant of 

 Crepis tectorum, showing fifteen chromosomes instead of the 

 normal eight. The four in black are the ring chromosomes, 

 whose abnormal separation has led to their reduplication ; 

 two are interlocked. Cf. origin of polyploidy, Ch. Ill (from 

 Navashin, 1930). 



of one clone that must originally have had six fragments, Langlet 

 found 252 divisions with the following fragment frequencies : — 



No. of fragments .345 6 7 89 10 

 No. of mitoses . i 21 40 148 23 14 i 3 



Similarly in Nicotiana the frequent loss of a fragment having an 

 abnonnal effect on flower colour gave patches of normal colour in 

 the abnormal (cf. Blakeslee, 1928, on Datura). 



Thus mitosis in some species seems to be better adapted to deal 

 with a wide range of size of chromosomes than in others and a 

 suitable mitotic mechanism is a second condition of the survival of 

 new types of chromosomes. 



