38 MITOSIS : THE CONSTANCY OF THE CHROMOSOMES 



(a) Each chromosome has centric constriction and perhaps 

 one or more secondary constrictions which are constant in position. 

 Both are usually intercalary, and it is doubtful whether they 

 can ever be terminal {cf. Lewitsky, 1931 ; Kaufmann, 1934 on 

 Drosophila ; D., 1936 b on Chorthippus). The alleged terminal 

 centric constrictions of grasshoppers and of the B chromosomes of 

 Zea Mays have been shown to be near the end, but not at the end 

 (D., 1936 and Fig. 102). A supposedly terminal nucleolar con- 

 striction gives the appearance of a " seta " (Kihara, unpublished). 



(b) Constrictions vary in the length of the lacuna they create. 

 This length is a characteristic property, although subject to variation 

 as an artefact, and, in part, as a natural condition. We may 

 therefore speak of a " long " constriction and a " short " constriction. 

 The real variation in appearance is general amongst very long 

 constrictions, especially those separating trabants (v. infra). Long 

 constrictions may be two or three times as long as the chromatid is 

 broad. At anaphase all constrictions are liable to be emphasised 

 by the tension between the centromere and the parts of the chromo- 

 some distal to the constriction. Centric constrictions are nearly 

 always short, but Aucuba and Aconitum are exceptional in this 

 respect {v. Fig. 79). 



(c) Prophase contraction of all sections of chromatids between 

 two constrictions proceeds until the chromatid reaches the diameter 

 characteristic of the race, unless it first reaches a spherical shape. 

 When the whole chromatid, or any section of it, is less than a 

 certain size, the characteristic breadth is not attained. In the first 

 case we have the spherical type of chromosome found in many 

 animals and in Muscari latifolium (Fig. 7 I B). In the second case, 

 merely a part of the chromosome is narrower than the rest (Fig. 7 1 C). 

 But it often happens that such a part is terminal and is separated 

 by a long constriction from the main chromosome, for long 

 constrictions are more frequent near the ends. We then have 

 what is called a " satellite " or trabanf. Occasionally this 

 condition is found in a small section intercalated in the middle of the 

 chromosome. We thenhave an " intercalary trabant " (Fig. 7 III N). 



This view that the trabant is distinguished from other parts of 

 the chromosome merely by its small size (D., 1926) has been dis- 



