368 AN INTRODUCTION TO MODERN GENETICS 



were true, one would have to consider the ends of the chromosomes as 

 permanent organs just as are the centromeres. There seems, in faa, to 

 be no case on record in which we can be certain that an end of a chro- 

 mosome has been knocked off so that a new chromomere forms the tip; 

 in all deficiencies which can be adequately tested, it is either proved, or 

 at least still possible, that the deficient section does not reach quite to 

 the end of the chromosome, and that the hypothetical end chromomere 

 is preserved. However, it would in most cases be very difficult to prove 

 rigidly that the end chromomere was missing, and the common occur- 

 rence of interstitial deficiencies may only indicate that they usually 

 arise by the breakage and rejoining of loops (the second mechanism 

 in Fig. 43) and not that the simple breakage (first mechanism) is 

 totally impossible. 



B. DEVELOPMENTAL CONSIDERATIONS 



The "Mendelian factor" was at first a purely abstract idea, defined 

 by the Mendelian laws which regulated its behaviour. The first sug- 

 gestion as to a possible material basis for it was the Presence and 

 Absence hypothesis, discussed and to some extent defended by Bate- 

 son. This was based on the observation that most mutant characters 

 are inherited as recessives and can be considered as in some way 

 defective when compared with the normal form. The hypothesis was 

 therefore suggested that the dominant gene was a material entity 

 responsible for the development of a character, while the recessive was 

 a simple absence of this entity. The hypothesis soon had to be modified 

 to account for multiple allelomorphism; it had to be admitted that 

 some recessives involved only a partial absence (i.e. a quantitative 

 diminution) of the dominant. A much more serious difficulty arose 

 when it was discovered that not only may the dominant gene mutate to 

 the recessive, but the recessive may mutate back to the dominant. It is 

 fairly easy to conceive how a gene or part of one may become lost, so 

 that a dominant becomes converted into a recessive, but much more 

 difficult to see how a gene can be produced ex nihilo to give a "back- 

 mutation." The final criticism of the theory, however, is that it is 

 inadequate to deal with the facts of X-ray induced mutation; it limits 

 speculation too narrowly to the consideration only of quantitative 

 changes, whereas it is necessary to consider quantitative changes as 

 well. The theory was essentially based on a direct translation of the 

 developmental phenomena of hypo-hyper-morphism into terms of 



